Terminal Railroad Revisited: Foreclosure of an Essential Facility or Simple Horizontal
Monopoly?
Author(s): David Reiffen and Andrew N. Kleit
Source: Journal of Law and Economics, Vol. 33, No. 2 (Oct., 1990), pp. 419-438
Published by: The University of Chicago Press
Stable URL: http://www.jstor.org/stable/725371
Accessed: 19/02/2009 12:37
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=ucpress.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Journal
of Law and Economics.
http://www.jstor.org
TERMINAL RAILROAD REVISITED:
FORECLOSURE OF AN ESSENTIAL
FACILITY OR SIMPLE HORIZONTAL
MONOPOLY?*
ANDREW N. KLEIT
Federal Trade Commission
and
DAVID REIFFEN
Capital Economics
I.
INTRODUCTION
rOLLOWINGthe Supreme Court's 1986 decision in Aspen Ski,' there has
been a renewed interest in the "essential facilities doctrine." The doctrine, as we understand it, is inconsistent with economic theory. Broadly
stated, the doctrine says that, if assets cannot be economically reproduced by another firm but are economically essential to all producers of
some good, then all producers of that good should have equal access to
the assets. The doctrine presumes that the firm controlling such assets
will not provide access equally to all firms. This problem is thought to be
particularly acute in cases where the firm owning the asset is one among a
group of competitors needing that asset to produce some other good.
The case that has been cited as "establishing" the doctrine and the
"classic essential facilities case"2 is United States v. Terminal Railroad
Association of St. Louis.3 Several recent articles4 on this case detail three
important propositions:
* The views and conclusions expressed here do not necessarily represent the
opinions of
the Federal Trade Commission (FTC), any of its members, the Bureaus of Economics and
Competition of the FTC, or Capital Economics. This article was written while we were both
staff economists at the FTC. We wish to thank Tim Brennan, Malcolm Coate, Alan Fisher,
Paul Godek, Randy Kroszner, Robert Lande, Marc Schildkraut, Mike Vita, an anonymous
referee, and the editors of this journal for their assistance with this article.
Aspen Skiing Co. v. Aspen Highland Skiing Corp., 472 U.S. 585 (1985).
2
Daniel Troy, Unclogging the Bottleneck: A New Essential Facilities Doctrine, 83
Colum. L. Rev. n. 67 (1983).
3
224 U.S. 383 (1912) and 236 U.S. 194 (1915).
4
Troy, supra note 2; Thomas Krattenmaker & Steven C. Salop, Anticompetitive Exclusion: Raising Rivals' Costs to Achieve Power over Price, 96 Yale L. J. 234 (1986); David
[Journal of Law & Economics, vol. XXXIII (October 1990)]
? 1990 by The University of Chicago. All rights reserved. 0022-2186/90/3302-0007$01.50
419
420
THE JOURNAL OF LAW AND ECONOMICS
1. A group of railroads constituting a subset of all railPROPOSITION
roadsenteringSt. Louis from the west jointly erected a railroadterminal.
2. The terminal was the only "feasible" terminal option
PROPOSITION
for rail trafficcoming into St. Louis from the east.
PROPOSITION
3. Certain railroads (nonowners of the terminal) were
either (a) denied access to the terminal,foreclosingthem from competing
with those railroadsthat did have access (the owners) to the terminal,5or
(b) requiredto pay higher prices for the terminalthan its owners paid.6
Fromthe standpointof economic theory, the conjunctionof these three
facts is surprising.Propositions 1 and 2 suggest that the terminalwas a
naturalmonopolyowned by verticallyintegratedfirms.Proposition2 also
suggests that the elasticity of technical substitutionwas low, so that between certainlocation inputswere used in fixed proportions.(One unit of
terminal plus one unit of "transportation"creates the final producttransportationto a specific St. Louis location.) The strongform of Proposition 3 then is curious, even paradoxical.Even if it were verticallyintegrated, why would a monopolistdeny access to any customer?Even the
weak form of Proposition3 is surprising.In the fixed proportionscase, a
monopolist at any level of production can realize the entire monopoly
profit. The monopolist accomplishes this by charginga price that, when
added to the competitive markupsat other stages, yields the monopoly
price for the finaloutput. So whethera monopolistis verticallyintegrated
or not, it has no incentiveto foreclose7or discriminateagainstotherfirms.
In short, economic theory suggest that there is no need for an essential
Gerber, Rethinking the Monopolist's Duty to Deal: A Legal and Economic Critique of the
Doctrine of Essential Facilities, 74 Va. L. Rev. 1069 (1988); Gregory Werden, The Law and
Economics of the Essential Facilities Doctrine, 32 S.L.U. L. Rev. 432 (1988); James Ratner,
Should There Be an Essential Facilities Doctrine? 21 U.C. Davis L. Rev. 327 (1988); Alan D.
Neale & D. G. Goyder, The Antitrust Laws of the United States of America (3d ed. 1980);
and Lawrence A. Sullivan, Handbook of the Law of Antitrust (1977).
5
Troy, supra note 2, at 452, writes, "Certain railroads were denied access to the terminal, foreclosing them from competing with those having access to the terminal." According
to Gerber, supra note 4, at 1079, "a consortium of railroad companies refused to permit a
competitor to enter the consortium and thereby denied the competitor access to the sole
switching station on an important railway line." Werden, supra note 4, at 444, writes,
"While Terminal Railroad involved a concerted refusal to deal, Otter Tail and Hecht (two
other essential facilities cases) do not."
6 In Krattenmaker and
Salop's view, supra note 4, at 234, "[T]he railroad operators
obtained a promise from the bridge owners (here the railway operators themselves) that the
bridge could be made available to other, non-owner, railroads on discriminatory terms."
Neale & Goyder, supra note 4, at 128, assert that "[t]he proprietary group had the power of
veto and to discriminate against any newcomer."
7
According to Tirole, "Very loosely, market foreclosure are commercial practices (including mergers) that reduce the buyers' access to a supplier and/or limit the suppliers'
access to a buyer" (Jean Tirole, The Theory of Industrial Organization 193 (1988)).
TERMINALRAILROADREVISITED
421
facilities doctrine, as firms do not have anticompetitive reasons to deny
access.
We examine this conflict. Our research indicates that the facts in this
case support economic theory: all customers were allowed to use the
terminal (and related facilities) on a nondiscriminatory basis. Further, the
record indicates that the Terminal Railroad Association acquired a monopoly via a series of horizontal acquisitions. Once the association acquired its rivals, its policies were consistent with the principles of monopoly pricing for its services. Hence, at least in this case, economic theory
and the facts are congruent. Thus, public policy toward similar relations
in the future can rely on a less exotic theory of economic behavior than
that suggested by the essential facilities doctrine.
II.
OFVERTICAL
THE BASICECONOMICS
RELATIONS
Welfare effects of vertical integration depend on whether the downstream production process is characterized by fixed or variable proportions.8 Fixed proportions exist if the proportion of the upstream product
in each unit of output sold by downstream (manufacturing) firms does not
change with changes in the price of the upstream input.
In the fixed proportions case, the price of final output cannot rise as a
result of a profit-maximizing monopolist integrating into manufacturing.
The logic is straightforward. Suppose that the upstream level is monopolized and the "downstream" industry is competitive so that the margin
earned at that level (the product price minus the input price) is equal to
the marginal production cost (net of input price). If the downstream margin is equal to the marginal cost of providing the service, the monopolist
chooses an input price so that the input price plus the competitive manufacturing margin yields a product price that maximizes profit (that is, the
marginal cost equals marginal revenue, where marginal cost includes the
cost of manufacturing). By choosing the appropriate input price, the upstream firm ensures that the final product price is at the monopoly level,
8
An extensive discussion of the economics of vertical relations can be found in Alan
Fisher and Richard Sciacca, An Economic Analysis of Vertical Merger Enforcement Policy,
6 Res. L. Econ. 1 (1984), and in Tirole, supra note 7, ch. 4, "Vertical Control." The original
research from which these discussions derive include Michael Waterston, Vertical Integration, Variable Proportions, and Oligopoly, 92 Econ. J. 129 (1982); Parthasaradhi Mallela &
Babu Nahata, Theory of Vertical Control with Variable Proportions, 88 J. Pol. Econ. 1009
(1980); and Frederick Warren-Boulton, Vertical Control with Variable Proportions, 82 J.
Pol. Econ. 783 (1974). There is also a literature on motivations for vertical integration related
to transaction cost savings, which we shall ignore here. A good overview of this literature is
Oliver Williamson, Assessing Vertical Market Restrictions: Antitrust Ramifications of the
Transactions Cost Approach, 127 U. Pa. L. Rev. 953 (1979).
422
THE JOURNAL OF LAW AND ECONOMICS
and thus the upstreamfirm receives the entire monopoly profit on that
product.
Suppose the monopolist decides to integrateforward.If the upstream
firm'scost of manufacturingis the same as the independentfirm, then its
total cost of production(here, upstreamplus downstreamcosts) is unaffected by integration. Hence, the output price that maximizes profit is
likewise unaffectedby integration.In this case, the internaltransferprice
(the implicitprice the upstreampart of the integratedfirmchargedto the
downstreampart) is identical to the price chargedto independentmanufacturers,and no additionalprofits are obtainedthroughintegration.
This leads to the basic conclusion that all the monopoly profitscan be
achieved by having a monopoly at one stage and that verticalintegration
does not affect the price of the final output. There are, however, some
complicationsto this analysis. First, if the monopolistis regulatedand is
thus prevented from chargingthe monopoly price, it will have an incentive to transferthe monopoly profit to an unregulatedentity at another
stage. Under these circumstances,a regulatedinputproducermay refuse
to sell to other downstreamfirmsin orderto establisha monopoly for its
affiliateddownstreamentity.9In this case, verticalintegrationcould lead
to foreclosure and higherprices.
In contrast, if the monopolistis unregulated,but the downstreammarket is not competitive (that is, the marginexceeds marginalcost), the
upstreammonopolistcan reduceoutputprice and increaseprofitabilityby
vertically integrating.Here vertical integrationreduces the monopolist's
cost of getting the product to the consumer. As with any monopolist,
lower cost will translateinto lower outputprice (albeitnot on a dollar-fordollarbasis). In this case, the manufacturermay chargedownstreamfirms
a price differentfrom the internaltransferprice.
A furthercomplicationoccurs if the monopolist'sinput can be used in
variableproportions.In this case, downstreamintegrationby the monopolist can increase its profitswhile the effect on the price of the finaloutput
is ambiguous.
The reason for the increased monopoly profitfrom integrationis that,
as the monopolist raises the price of input, downstreamunintegrated
manufacturerssubstitute away from that input. This leads to production
cost above the level that would prevail if the monopolist's input were
9 This may explain the observed behavior in several essential facility cases, including
MCI Communications Corp. v. AT&T, 708 F.2d 1081 (7th Cir.), cert. denied, 464 U.S.
891 (1983), and Otter Tail Power Co. v. United States 410 U.S. 366 (1973). This point is
discussed in Timothy J. Brennan, Why Regulated Firms Should Be Kept out of Unregulated
Markets: Understanding the Divestiture in United States v. AT&T, 13 Antitrust Bull. 741
(1987), as well as by Gerber, supra note 4, at 1087-88.
TERMINALRAILROADREVISITED
423
availableto unintegratedmanufacturersat its marginalcost. Hence, if the
monopolistcould enter manufacturingwith the same cost function as an
independentmanufacturer,its total cost of getting the final product to
marketwould be less than that of an unintegratedmonopolist.
This incentive may induce the upstreammonopolist to fully integrate
downstreamand replace all of the independentproducers.If the upstream
firmhas available to it the same cost function as an independentdownstreamindustry, it will be in its interest to become the sole downstream
producer.The effect of this complete verticalintegrationon productprice
is ambiguous.On one hand, the actualcost of manufacturingfalls because
integrationinduces a more efficient input mix. All other things being
equal, this will increase the outputof that firm(relativeto an independent
one) and reduce output prices. On the other hand, the ability of downstream firms to substitute away from the monopolist's input no longer
constrainsthe integratedmonopolist, and it can thus more effectively use
its monopolypower by raisingthe (implicit)inputprice. The net result on
the price of final output of these two effects is ambiguous.
If the monopolistdoes choose to integratefully downstream,the input
price chargedto independentmanufacturers,yieldingthe monopolistthe
same profitas it earns on its own output, is sufficientlyhigh to discourage
independentfirmsfrom producing.In this sense, independentproducers
are "foreclosed" from the market,althoughthere is no refusalto deal.10
The theory of vertical relationshipsreviewed here indicates that the
monopolistcan fully exploit its marketpower withouta refusalto deal. In
reality, however, firms do refuse to deal. For example, it is common for
manufacturersin certain industriesto sell their productsthroughdealers
havingexclusive territories.In effect, producersrefuse to deal with all but
one purchaserin a given geographicarea. A commonexplanationfor such
actions is that they representan efficient way to marketproducts.1
10
While it may be argued that this represents a de facto refusal to deal, the monopolist is
willing to sell at a price that meets this criterion, and if the independent producer has
sufficiently lower costs than the monopolist, it would be willing to buy at that price. A
somewhat more complex case occurs if diseconomies of scale in production are sufficient to
make it unprofitable for the monopolist to become the sole downstream producer. In this
case, the input monopolist may partially integrate downstream and continue to sell to other
producers. Here again, the integrated firm will have lower costs than an unintegrated one
and will tend to increase output. The converse is that the input price charged to independent
producers will rise, causing them to contract output. As in complete integration, the effect
on price is ambiguous. (In at least one case, price will rise as a result of partial integration.
Herman C. Quirmbach, Vertical Integration: Scale Distortions, Partial Equilibrium, and the
Direction of Price Change, 101 Q. J. Econ. 131 (1986), has shown that, if all costs are
variable, and the downstream industry is initially in a zero-profit equilibrium, then price
necessarily rises.)
1 See Benjamin Klein & Kevin M. Murphy, Vertical Restraints as Contract Enforcement
424
THE JOURNAL OF LAW AND ECONOMICS
Anotherefficiency-motivatedreason for a refusalto deal mightoccur if
both the upstream and downstream markets are (unregulated)natural
monopolies. Here, if separate firms own the upstreamand downstream
inputs, there is a potentialfor a successive monopolyproblemto emerge;
as the two monopolists attempt to individuallymaximizeprofits, the result is lower total profit (and consumer welfare) than if one firm owned
both monopolies.12Thus, if a vertically integratedfirm refuses to deal
with a downstreamrival, economic theory indicatesthat such a refusalis
likely to be efficient and should not be hinderedby antitrustlaws.13
Essential facilities cases would seem, almost by definition,to approximatethe fixed proportionscase. The "essential" aspect impliesthatusing
the essential input is the only economical way of producingthe output.
Troy14writes that a facility is essential if the end product cannot be
(economically)producedwithoutusing the facility. Gerber'5goes further,
as he assumes throughouthis article that fixed proportionscharacterize
all essential facilities cases. Gerber'sassumptionis strongerthan Troy's
in that variable proportions can still exist even if the good cannot be
producedwithout some amountof the monopolizedinput. Nevertheless,
fixedproportionsseem to characterizeproductionin manyessentialfacilities cases.
The situationin TerminalRailroadprovidesone example. If the inputis
viewed as St. Louis-area bridges, for certain origin/destinationpairs
there were no close substitutes, and fixed proportionsresults apply.'6
Shippers in other locations could substitute away from St. Louis-area
Mechanism, 31 J. Law & Econ. 265 (1988); and Howard P. Marvel, Exclusive Dealing, 25 J.
Law & Econ. 1 (1982) for extended discussions of the efficiency rationale. The arguments
made in these articles are traceable to Lester Telser, Why Should Manufacturers Want Fair
Trade? 3 J. Law & Econ. 86 (1960).
12
See Joseph J. Spengler, Vertical Integration and Antitrust Policy, 68 J. Pol. Econ. 347
(1950). Fishman v. Wirtz, 807 F.2nd 520 provides an interesting illustration. The owner of
Chicago Stadium (who was bidding to buy the Chicago Bulls basketball team) allegedly
refused to offer a stadium lease to a competing bidder for the team. This insured that the
stadium owner won the bid, avoiding a successive monopoly problem. It is also possible that
a desire to avoid future postcontractual opportunism motivated the stadium owner (as in
Benjamin Klein, Robert G. Crawford, & Armen A. Alchian, Vertical Integration, Appropriable Rents, and the Contracting Process, 21 J. Law & Econ. 297 (1978)). A type of postcontractual opportunism may also describe the events in Aspen Ski, supra note 1.
13 A similar point is made by Gerber, supra note 4, at 1085-86.
14 See
Troy, supra note 2, at 459.
15 See Gerber,
supra note 4.
16
The results in Waterston, supra note 8, indicate that, if proportions are "close to" fixed
(that is, there is little opportunity for substitution), the results are similar to those when the
proportions are fixed.
TERMINALRAILROADREVISITED
425
bridges to other Mississippi River bridges without a significant increase in
costs (see Section IV for details). For these customers, St. Louis bridges
represented a "variable proportion input," and the association had little
ability to increase price. This suggests that, whatever market power the
association held, it could be most easily expressed by charging a price for
river crossing/terminalling services that would result in monopoly prices
for shipping freight. In this case, we would anticipate that the price for
these services would be the same to association members and nonmembers.
Terminal Railroad is interesting in another respect. The economic analysis of vertical integration is well developed but remains largely at a
theoretical level. The type of transactions described above are usually
internal to a particular firm. Thus it is difficult or impossible for an outside
observer to determine if the reality of vertical integration fits the economic theory. In this case, however, we can determine what price was
charged and what type of discrimination, if any, occurred.
III.
HISTORYOF THETERMINALRAILROAD
CORPORATE
ASSOCIATION OF ST. LOUIS
Evaluating the history of the Terminal Railroad Association is important in understanding the economic environment that led the attorney
general to bring suit against the association. As the following discussion
illustrates, the activity that the government found objectionable was horizontal in nature. In particular, via a series of horizontal acquisitions, the
association gained a monopoly over methods of shipping freight across
the Mississippi at St. Louis. The pricing policy of the association simply
reflected this monopoly.
In 1874, the Eads Bridge crossing the Mississippi River at St. Louis was
completed. At the same time, a tunnel was constructed connecting the
bridge to the valley of the Mill Creek, where the railroads located on the
Missouri side of the river were situated. Separate firms, the Union Railway and Transit Company of St. Louis and the Union Railway and
Transit Company of Illinois, built tracks to connect these facilities to
railroads entering the St. Louis and East St. Louis areas. These firms also
provided the locomotives necessary to transport freight across the river.
These facilities proved inadequate (at least according to defendants). In
1880, the Terminal Railroad of St. Louis (which is distinct from the Terminal Railroad Association of St. Louis) was incoporated for the purpose of
"provid[ing] the most ample and convenient connection and accommodation and terminal facilities in St. Louis for all railroads now entering or
426
THE JOURNAL OF LAW AND ECONOMICS
hereafter[entering]the same." 17 They then leased these terminallingfacilities to the companies operatingthe bridge and the tunnel. One year
later, two railroads,the Wabash,St. Louis, and Pacific(the Wabash)and
the Missouri-Pacific,became joint lessees of the bridge and tunnel and
sublessees of the terminallingfacilities.
By 1889,Jay Gould had acquiredsufficientstock in both of these railroads to exercise control over them. In that year, Gould promoted an
agreementbetween these two railroadsand four additionalrailroadsthat
also had terminalsin St. Louis, creatingthe TerminalRailroadAssociation of St. Louis (the association). Throughthis arrangement,the association acquired the properties of the TerminalRailroadof St. Louis, the
depots on both sides of the river, and the assignmentof the lease (previously held by the Missouri-Pacificand the Wabash) to the bridge and
tunnel.
In 1886, an act of Congress authorizedthe constructionof a second
bridgeat St. Louis. One provision of the act prohibitedany person who
was a stockholder, director, or manager in any other bridge over the
Mississippi from becoming a stockholder, director, or manager in the
second bridge.18At trial, the former governor of Missouritestified that
the specificintentof this provisionwas to ensurean independentcompetitor to the Eads Bridge,19suggestingthat priorto the constructionof the
second bridge,the owners of the Eads Bridgewere in a position to charge
monopoly prices.
The second bridge, known as the MerchantsBridge,opened on June 1,
1890.20The company that owned the bridge, the St. Louis Merchants
BridgeTerminalCompany,also securedor builta series of smallrailways
in Illinois and Missouri. Accordingto the government,these railways, in
combinationwith the MerchantsBridge, constituted a system that provided "branches, switches, and depots, so as to enable it to conduct
interstateand internationalcommerce across the Mississippi."21As evidence for this, the governmentproduceda tariffschedulethat showed the
MerchantsCompanyposted rates for all area railroadson either side of
the Mississippi.The tariffschedule, accordingto the government,demon17
Appellees' Statement and Abstract (hereafter cited as ASA) U.S. v. Terminal Railroad,
224 U.S. 383, at 11.
18 Abstract of the
Pleadings and Evidence (hereafter cited as ABS) U.S. v. Terminal
Railroad, 224 U.S. 383, at 23.
19 Statement and Brief of the Attorney General (hereafter cited as Brief) U.S. v. Terminal
Railroad, 224 U.S. 383, at 19.
20
ABS, supra note 18, at 4.
21 Brief,
supra note 19, at 51-52.
TERMINALRAILROADREVISITED
427
stratedthat all railroadsconnected to the Merchantsterminalsystem.22In
short, the governmentarguedthat the St. Louis MerchantsBridgeTerminal Company constituted a competitive system to that of the Terminal
RailroadAssociation.
By 1893,the provisionof the act prohibitingindividualsowninga share
in the TerminalRailroadAssociation from owning any partof Merchants
Bridge had been deleted in some "mysterious manner."23In that year,
the members of the association acquired the right in perpetuity to use
MerchantsBridge and its terminallingfacilities. In exchange for this, the
association guaranteed$3.5 million of the MerchantsBridge Company's
bonds and purchased 4,384 shares of stock. According to the association,24this infusionof cash and credit preventeda foreclosureof some of
the MerchantsBridge Company's assets. At the same time, the association acquired 13,416 additionalshares of stock, thus securingcontrol of
the MerchantsBridge Company.25
Althoughthere were only two St. Louis-area bridgesover the Mississippi at this time, the association faced competitionfrom other sources,
primarilyferry companies. The largest of these was the Wiggins Ferry
Company.It owned several miles of riverfronton the Illinois shore opposite St. Louis. On this riverfrontproperty, the company built switching
yards and other terminallingfacilities.26It also owned the stock of the
company operatingthe East St. Louis ConnectingRailroad,which connected with the various railroadlines enteringIllinois towns such as East
St. Louis, East Carondelet,Madison, and Venice.27Similarly,the Wiggins Companyoperated facilities on the St. Louis side of the river, and
throughthese facilities it was able to connect with railwaylines terminating on that side of the river.28With these facilities (and two dedicated
ferries), the Wiggins Company was able to ferry 1,200 railroads cars
across the river each day.29 The government argued that the Wiggins
system, like the Merchantssystem, reached the same competitive territory in the center of St. Louis and in Illinois as the association's system
and reached "practicallythe same railroadsin the two states."30Hence,
22
23
24
25
26
27
28
29
30
Id. at 81.
Id. at 53.
ASA, supra note 17, at 20.
Id. at 20; and Brief, supra note 19, at 53.
224 U.S. 385; and ABS, supra note 18, at 73.
ABS, supra note 18, at 20; and Brief, supra note 19, at 46.
224 U.S. 385, supra note 26.
Brief, supra note 19, at 82.
Id.
428
THE JOURNAL OF LAW AND ECONOMICS
the government concluded that the Wiggins and Merchant companies
were independent instrumentsof interstate commerce, competing with
the TerminalAssociation.
In 1902, the Chicago, Rock Island, and Pacific Railroad(the Rock Island), then not a member of the association, attemptedto purchase the
Wiggins company.31According to the government, this engendered a
takeover battle for Wiggins stock between the association and the Rock
Island.This battleeventuallyresultedin the acquisitionby the association
of 9,500 of the 10,000outstandingshares in the Wigginscompany.32The
total expenditureon Wigginsstock was $7,426,356,or about$782for each
of the 9,500 shares.33As part of the ultimatesettlement, the Rock Island
and seven other railroadswere admittedto the association. The association's holdings in the Wiggins company were divided equally among 13
association memberrailroadsand the Pennsylvaniacompany(which was
affiliatedwith the fourteenthassociation member).34
Two other ferry/railroadcombinations provided rail connections between the two sides of the Mississippi.One was createdby the Pennsylvania railroad,which acquiredsome existing trackandexpandedthe system
to the eastern bank of the Mississippi. This "belt" road, known as the
Conlogueroad, connected with the Pennsylvania'sown tracks, as well as
with every other railroadin the city of East St. Louis, and reachedevery
terminallocated in that city. The southern terminusof the road was in
East Carondelet,Illinois (a few miles south of St. Louis), wherethere was
a ferry dock owned by the Missouri-Pacificrailroad.Cars were transferred directly from the rails of the Conlogue road onto the ferry, and
from the ferry directly to the tracks of the Missouri-PacificRailroadin
South St. Louis. The ferry handled200-300 cars per day duringits period
of operation.
In 1902,the associationpurchasedthe Conlogueroadfromthe Pennsyl31 Id. at 33.
ABS, supra note 18, at 33.
33 Id. at 110. The various filings provide contradictory evidence on the price actually paid
for the shares. For example, the government brief claims the price was either "greater than
$725" (id. at 56) or $1,500 (id. at 89). Also, a witness claims that he got $500 per share, which
was more than the stock was worth (id. at 110). Our interpretation of these claims is that, as
time went on and the fight for the stock of the Wiggins Ferry intensified, the offer price rose.
Early sellers received approximately $500 for their shares, later sellers $1,500. The average
price paid was $782, while the government claimed that they were "worth" only $300 (Brief,
supra note 19, at 89), although it is not clear whether this figure represents book value,
replacement value, or some other concept of worth. The same ambiguity is true of the
government's use of "worth" in regard to other association acquisitions. The association's
1914 annual report shows a book value for the Wiggins shares of $743.67.
34 ASA, supra note 17, at 71.
32
TERMINALRAILROADREVISITED
429
vania company for $1.2 million. (The governmentgave no estimate of
either the replacementor book value of the Conlogueroad.) At the same
time, an affiliate of the Pennsylvania railroadbecame a member of the
association. Subsequent to this purchase, the rail line ceased delivering
cars to the ferry service at East Carondelet,and the ferry ceased operations.35The association boughtanotherferry operator,the InterstateCar
Transfer Company, for $600,000 (the government alleged it was worth
only $225,000).36The governmentclaimed that both of these companies
competed with the association.37
Of course, the TerminalAssociation's policy of buyingup existingferry
companieswould encourage new firmsto enter (or threatento enter) the
ferry business in St. Louis.38There is, however, no court recordof additional firms entering. Entering this industry requiredriverbankland on
both sides of the Mississippi that would make connections to railroads
possible. Apparently,the WigginsCompanyalreadyowned much of the
suitable land. Obtaining the required land could be made even more
difficultby the necessity of buying it from several owners, each of which
could ask for a sizable portion of the available profits. Thus, given the
nature of the task and the fact that there is no record of new entry, it
would appearthat there were sizable difficultiesin establishingan additional ferry company.
The toll bridgeacross the Mississippiclosest to St. Louis was at Alton,
Illinois. While this bridgewas not an association property,its ownership
consisted of eleven railroads, ten of which were association members.
This suggests that the association could have effectively prevented the
Alton bridge from competing for traffic with the association. In fact,
accordingto the managerof the Alton bridge, after the acquisitionof the
bridge by the eleven firms, the transportationprice charged for a hundredweightof coal rose from eight cents to thirty cents.39
IV.
THEBRIDGE
ARBITRARY
By virtue of its leases with the Merchantsbridgeand its acquisitionof
the three ferry companies, the TerminalRailroadAssociation controlled
all feasible channels of transportationacross the Mississippiinto and out
Brief, supra note 19, at 84-87.
ABS, supra note 18, at 77.
37
Brief, supra note 19, at 109.
38 Standard Oil faced a similar
problem of buying up new entrants in order to protect
market power during this time period. See John McGee, Predatory Price Cutting: The
Standard Oil (N.J.) Case, 1 J. Law & Econ. 137 (1958).
39 ABS, supra note 18, at 99.
35
36
430
THE JOURNAL OF LAW AND ECONOMICS
of St. Louis. Priorto these acquisitions,the item most often shippedinto
St. Louis from the east was coal,40a productproducedprimarilyeast of
the Mississippi.41The association's annual reports indicate that almost
half of the westbound trafficacross the Mississippi at St. Louis in 1896
was coal.42Morethan 70 percentof the coal used in St. Louis in 1907was
mined in Illinois within thirty miles of East St. Louis.43It would have
been extremely difficult to route coal from these areas into St. Louis
without using association properties. This meant that St. Louis coal
buyerswere limitedin their abilityto shift patternsof tradein response to
a higherprice for association services.
In the one year for which data are available(1895),the physical volume
of eastbound goods transitingthe association's tracks originatingin St.
Louis was less than 40 percent of the westboundtrafficterminatingin the
city.44Judgingfrom the complaints of the plaintiffs witnesses,45it appears that the bulk of the goods travelingeastboundconsisted of grain,
flour, and other milled products and some manufacturedgoods.46Merchants sellingthese products,like the buyersof coal, were limitedin their
ability to shift patternsof trade when faced with an increase in the price
chargedto cross the river.47
Buyers and sellers in other cities west of the Mississippi, however,
were not "captive" to the St. Louis bridges. Many other toll bridges
spanned the Mississippi, including those at Keokuk, Davenport, and
Dubuque, Iowa; Hannibal and Louisiana, Missouri; Quincy, Illinois;
40
Brief, supra note 19, at 75.
Only 11.5 percent of the U.S. production of coal in 1900 was mined west of the
Mississippi, and this share had declined to 10.5 percent by the year preceding the Supreme
Court decision. Pennsylvania, Illinois, Kentucky, Alabama, Ohio, and Indiana were (in
order) the largest producing states (Northern Illinois Coal Trade Association, Production of
Coal, Bituminous and Anthracite, Years 1800 to 1974 Inclusive, by States and Producing
Districts and the United States 1976).
42 1896 annual
report of the Terminal Railroad Association of St. Louis (Con P. Curran
1896).
43 Brief, supra note 19, at 59.
44 1895 annual report of the Terminal Railway Association of St. Louis (Con P. Curran
1895), table V.
45 ABS, supra note 18, at 88-101.
46 The most viable alternative for manufacturers seemed to be
locating on the Illinois side
of the river. In this vein, prosecution claimed that the growth of certain manufacturing cities
in Illinois (Granite City, Madison, and East St. Louis) was "almost entirely due" to the
Bridge Arbitrary (discussed below) (Brief, supra note 19, at 25). This indicates that, for the
products manufactured or processed in St. Louis, it was not practical to reroute shipments
or find new markets that could be reached without using association facilities.
47 ABS, supra note 18, at 98, the president of a flour mill said that 85-90 percent of his
shipments were eastbound. Because of this, the Bridge Arbitrary had a "good deal" to do
with the mill moving from St. Louis to Alton, Ill.
41
TERMINALRAILROADREVISITED
431
Memphis, Tennessee; and Minneapolis/St. Paul, Minnesota.48Freight
travelingto Kansas City or points west could easily be routedaway from
the St. Louis crossings.
These competitive conditions explain the pricingpolicy of the association. The association had no marketpower over shipmentsgoing from St.
Louis to the west. Since numerousother crossings options existed, the
associationhad no marketpower over shipmentsfrom east of the Mississippi to points beyond St. Louis (or vice versa). Therefore, the only
marketpower available to the association was on shipmentsgoing into
and out of St. Louis to points on the eastern side of the Mississippi.
This marketpower was exercised throughthe "BridgeArbitrary."The
Bridge Arbitrarywas a toll of two cents per hundredweightfor goods
travelinginto or out of St. Louis.49There was no analogous charge for
goods passing throughSt. Louis over one of the bridges in either direction.50This aspect of the asssociation's pricingwas a centralpremise of
the government'scase.5
Even the association's marketpower over St. Louis trafficwas limited.
Trafficto andfrom southernstates52did not have to pay the arbitrary.The
apparentreason was that a directconnectionto these areasexisted via the
St. Louis and San Francisco railroadusing the Memphis Bridge.53As
productsgoing to this region had a good substituteavailable,the association's marketpower was diminished,and the associationcould not charge
the arbitrarywithout losing most or all of this business.
V.
A.
THEISSUESATTRIAL
History of Court Proceedings
In 1904 the supreme court of Missouri heard a case brought by the
Missouri attorney general arguingfor the dissolution of the merger be48 Commissioners Official
Railway Map of Missouri, Higgins & Co. (undated, but "correct to June 1, 1902"), also Centennial American Republic and Railroad Map of the U.S. and
the Dominion of Canada in Andrew M. Modelski, Railroad Maps of North America: The
First Hundred Years (1984).
49 Brief,
supra note 19, at 100.
50 We found no evidence in the court records that the association was constrained in
its
pricing behavior by any government regulation. Although the Interstate Commerce Commission was established in 1887, it was not given the power to enforce maximum rates until
1906, and it did not have the power to enforce minimum rates until 1920. See Thomas Ulen,
The Market for Regulation: The ICC from 1887 to 1920, 70 Am. Econ. Rev. 306 (1980).
51 See note 59, infra, for details.
52 Specifically, the area south of the Ohio River, east of the Carolinas and
Georgia, and
extending to the Gulf of Mexico (Brief, supra note 19 at 31).
53 Map of the "Frisco" line in
Modelski, supra note 50.
432
THE JOURNAL OF LAW AND ECONOMICS
tween the association and the MerchantsBridge.54The case was brought
forth under Section 12 of the Missouriconstitutionthat forbademergers
between "parallel" (competing)railroads.
On a four-to-threedecision, the Missouri court held for the Terminal
Association. According to the majority, the provision in the Missouri
constitutioncalled for what we would now term a "rule of reason" standardwith regardto railroadmergers. Since the majorityof the court felt
that there were importantefficiencies in the merger(fromcombiningthe
switching facilities of the two systems), it refused to dissolve the arrangement.
In 1905, the U.S. attorneygeneral broughtsuit againstthe association
in U.S. DistrictCourtfor Missouri.The court split two to two on the case.
Since an evenly divided court does not generally write opinions, we do
not know the reasoningof the judges involved at the district level.
The case was appealeddirectlyto the SupremeCourt.In 1912,by a sixto-zero vote, the Courtruledthe association, as constituted,to be illegal.
The Courtorderedthe association to permitother railroadsto be able to
become association members, serve nonmemberson equal terms, and
eliminatethe Bridge Arbitrary.55Only if the association refused to meet
these terms did the Courtcall for the remedythat would have gone to the
heart of the monopoly problem, the dissolution of the association.56
B.
Efficiency versus Monopolization as a Rationale for the
Association's Acquisitions
As noted above, a central contention of the federalgovernment'scase
was that the purpose of the association was to eliminatecompetitionby
acquiringMerchantsBridge and the three ferry companies. The government argued that the Wiggins and Merchants systems were "parallel"
lines to those of the Terminal Association and that the acquisitions
State v. Terminal Association of St. Louis 182 Mo. 284 (1904).
In the 1914 annual report of the Terminal Railroad Association of St. Louis (Con. P.
Curran 1914), we found that one additional road had joined the association, bringing the
number of members to fifteen. The Supreme Court decision refers to a total of twenty-four
railroads serving St. Louis.
56 While
eliminating the Bridge Arbitrary would be expected to result in lower prices for
St. Louis freight, it would also be expected to cause higher crossing prices for other cargoes,
as the association adjusted the method by which it took advantage of its market power. The
welfare effects of proscribing price discrimination such as the arbitrary are indeterminate.
There is a long line of literature in this area, starting with Joan Robinson, The Economics of
Imperfect Competition (1933), at 188-202, and extending at least to Michael L. Katz, The
Welfare Effects of Third-Degree Price Discrimination in Intermediate Good Markets, 77
Am. Econ. Rev. 154 (1987).
54
55
TERMINALRAILROADREVISITED
433
amounted to monopolizationof interstate commerce. In its brief,57the
governmentclaimed that the monopolizationled to a variety of deleterious effects, including(i) poor service, (ii) deterringentry at the terminal/
rivercrcssing level, and (iii) high prices (linkedto the BridgeArbitrary).58
The defense argument,in effect, amountedto the claim that providing
terminallingand switchingservices was a naturalmonopoly. The association asserted that use of the combinedfacilities of all five companieswas
necessary to service all locations in St. Louis with maximumefficiency. A
subsidiaryclaim was that the three majorsystems did not truly compete
since they served separate areas. Further,the combinationof assets led
the association to make investmentsin enlargingand connectingfacilities
to enhance the movement of trains. Hence, the combinationof assets led
to greatercompetition,as each individualrailroadcould, as a resultof the
acquisition, provide service to any individualcustomer.
C.
Profitability of the Association
As detailed in Section IV, one apparentaim of the TerminalRailroad
Association was to eliminatecompetitionfor rail services into and out of
St. Louis. Given this, one would expect that the association's actions
would result in high profitsfor it and its members.Reachingthis conclusion with the availablefacts, however, is not a simple matter.
The only informationwe have available to use on profits comes from
the association's accounting figures. Use of accountingdata to analyze
the existence of monopoly profits has been subject to criticism in the
academicliterature.59Accountingdata serve a variety of purposes, many
of which have little to do with true economic profitability.These data are
57
58
Brief, supra note 19, at 104-5.
The government'schargescan be summedup in these three categories.The actuallist
included (i) causing delays in transit and delivery; (ii) deprivingthe city of St. Louis of
adequatefreight facilities; (iii) exercising favoritismto St. Louis TransferCo. (a wholly
owned subsidiaryof the WigginsCo.); (iv) using the "GuaranteeAgreement"(discussed
below); (v) renderingindependentfreight stations impossible;(vi) establishinga "Bridge
Arbitrary"hostile to business interestsof St. Louis shippersand manufacturers;(vii) handlingcoal trafficto St. Louis's detriment;and (viii) arbitrarilyfixingand maintainingfreight
ratesto St. Louis's detriment.The firsttwo items suggestspoor service, the fourthandfifth
suggestan attemptto preventcompetitionwith the association,and the last threedeal with
the effect of the association'spricingpolicy (thatis, price discriminationagainstSt. Louis).
The thirditem suggests that haulingservices were used in variableproportionwith terminallingor railway service and, hence, that a lower price (or, equivalently,higher quality
service)was chargedto the association'sown haulingcompany.It shouldbe notedthatonly
a small percent of the association's revenue was derived from the transfercompany, and
hence this issue was not centralto this case.
59 See, for example, FranklinM. Fisher& JohnJ. McGowan,On the Misuseof Accounting Rates of Returnto Infer MonopolyProfits,73 Am. Econ. Rev. 82 (1983).
434
THE JOURNAL OF LAW AND ECONOMICS
often calculatedusing depreciationschedulesand cost-allocationschemes
that may not bear a strong resemblanceto reality and may thus produce
misleadingrate-of-returnestimates.
Despite this caveat, a review of the Terminal Association's annual
reportsfor several of the years in question indicatesthat there is at least
some reason to believe that the associationearnedreal economic profit.60
From calendaryear 1893 to 1897, the association had average profits of
$80,000 per year on average revenues of $1.758 million (4.5 percent).61
Accountingprofitabilityincreased to $313,000(13.9 percent of revenue)
duringthe periodfrom 1898to 1902.The firstfew years of data following
the aquisitionof the Wigginscompany showed higherprofits.From fiscal
year (July to July) 1906 to 1909, the association averaged profits of
$567,000on revenues of $2.911 million (19.5 percent). From fiscal year
1910to 1914the association averaged profitsof $257,000on revenues of
$3.107 million (8.3 percent). (The annualfiguresfor 1903, 1904, and 1905
were not available.)
These profitfiguresdo not include the fees the association was paying
its member roads. Prior to 1903, the association reports indicate that it
paid $550,000per year (an amountequivalentto about 27.5 percentof the
association's revenues) to the Wabash and the Missouri Pacific roads
(originalpartnersin the association)for use of tracksand the tunnelin the
northernpart of St. Louis. After 1905, the reportsindicatethat rental of
$666,900per year (22.1 percent of revenues) was paid to these roads for
use of the track and tunnels. It may well be that these were justified
expenses. Given the largeamountof the association'scosts accountedfor
by these outlays, however, it would seem likely that the Wabashand the
MissouriPacificsiphonedoff partof the association'sprofitsas partof the
originalagreementestablishingthe association.
D.
Exclusionary Practices
The standardinterpretationin this case is that railroadsthat were not
members of the association either paid more for, or were completely
foreclosed from, using terminals. The genesis of this interpretationis
unclear. Perhaps it derives from the Supreme Court's remedy, which
focused on the association's abilityto exclude competitors,as opposed to
60 Of course, the Terminal Association denied that its fees were too high. According to
them, "the charge of extortion is absolutely without support in this testimony. There was
not even an effort to support it, and on the other hand, the evidence is conclusive that the
charges are based on cost of operation, proper maintenance and fixed charges" (Appellees
Statement and Brief (hereafter cited as APP), U.S. v. Terminal Railroad, 224 U.S. 383, at
25).
61
The annual reports indicated that each member of the association owned an equal
amount of shares and thus shared equally in any dividends paid by the association.
TERMINALRAILROADREVISITED
435
the effects of the association's pricingpolicy.62At no point, however, in
the government's brief or in any other document, does the government
make any accusation of foreclosure.63On the contrary, the government
averredthe reverse. For example, Joseph Ramsey, a formergeneralmanager of the association and a governmentwitness, testifiedthat all terminals were availableto all roads on exactly the same terms.64Additionally,
the Departmentof Justice claimed that the association "compel[ed] . . .
all other railroadcompanies to use the propertyof the Association,"65a
positionthat seems to be the opposite of claimingforeclosure.The association made the lack of favoritisma central theme in its defense.66
There are three senses in which some notion of foreclosure played a
part in the case. First, the members of the association had the ability,
which was never exercised, to exclude railroads from the use of the
terminals.The economics of vertical integration(discussedabove in Section II) suggests that the monopoly price could have been set unilaterally
by the association, and hence the association did not need to refuse to
deal with nonowners to achieve its objectives.
Second, it took a unanimousvote of association membersto admitnew
members. Certainrailroadsthus may have been kept out of the association. Of course, if we accept the government'scontentionthat the association earned monopoly profits, it is not surprisingthat the association
would not want to admitnew members.New memberswould not increase
the total profitabilityof the association. Hence, unless new memberspaid
an admissionprice at least equalto (the presentvalue of) theirshareof the
monopoly profits, incumbent members would be made worse off. Of
course, new members would be unwillingto pay an amount exceeding
their shareof monopolyprofitssince admissionto the associationwas not
a necessary condition to use its facilities.
Third, each railroadpromisedto use only TerminalAssociation facilities to cross the river (the "GuaranteeAgreement"). This made entry
difficultfor any potentiallycompetingbridgeor ferry since the competitor
could not obtain the business of the fourteenrailroadsin the association.
In essence, the government argued that integrationby the "upstream"
(terminalling)monopolistwith firmsat the "downstream"(railroad)level
62
Remedies that have little to do with the problem at hand are not that uncommon in
antitrust proceedings. See Kenneth Elzinga, The Antimerger Law: Pyrrhic Victories? 12 J.
Law & Econ. 43 (1969); and Robert A. Rogowsky, The Economic Effectiveness of Section 7
Relief, 12 Antitrust Bull. 187 (1986).
63
As pointed out in notes 5 and 6 supra, this fact seems to have been overlooked in most
studies of the essential facilities doctrine. One exception is Ratner, supra note 4, at 337, who
notes that "no denial of access was alleged or shown (in the Supreme Court record)."
64 ASA, supra note 17, at 110.
65
ABS, supra note 18, at 35.
66 For example, APP, supra note 60, at 29, 50-51.
436
THE JOURNAL OF LAW AND ECONOMICS
foreclosed access to the downstreaminputfrom potentialcompetitorsat
the upstreamlevel. Withoutsuch access, the upstreamfirmcould not sell
its service. The government'sargument,therefore, seems to be that the
"GuaranteeAgreement" forced an entrant to enter at both levels and,
perhapseven then, at an inefficientlysmall scale. This idea of foreclosure
has had a long and controversialhistory in the antitrustliterature.67
This argumentseems curious since several railroadswere unaffiliated
with the association. Further, even associations members could be induced to contract with the entrant if the combined enterprise were
profitable.The fact that no complainantcame forth claimingit had been
foreclosed from establishing a terminalby the agreementsuggests that
foreclosurewas not a real effect of the agreement.
Regardlessof the plausibilityof this version of a foreclosureargument,
if exclusion of this type were possible, it would underminethe role of the
essential facilities doctrine. If the "GuaranteeAgreement" had an anticompetitive effect, it must have been througheliminatingthe threat of
competitionat the terminallinglevel. The basis of the doctrine, however,
is that the monopolizedinput should be made availableto all because it is
"commercially impractical"for a competitor at the upstream level to
reproducethe downstreamasset. If entry at the upstreamlevel were a real
threatto the monopolist, then clearly it was not "commerciallyimpractical" to replicatethe association's facilities.68
In sum, the first notion of foreclosure was not a cause for concern.
Monopoly prices for terminal services were already in effect, and the
association had no incentive to exclude anyone. The second type of exclusion is not an antitrustissue at all. Forcingthe associationto accept all
railroadsas membersmay have redirectedsome profits,but it would not
have changed output or prices to consumers. The final sense of foreclosure may or may not have had economic merit,but it does not supportthe
essential facilities doctrine.
67
For example, Phillip Areeda, Antitrust Analysis: Problems, Test, Cases 675 (2d. ed.
1974), argues that integration by the upstream firm forces a potential upstream entrant to
enter on both stages. "The additional capital, expertness, and facilities required to enter
simultaneously on both levels will obviously increase the difficulty of entering." As discussed in Section II supra, however, vertically integrated firms do have important incentives
for dealing with firms that compete with them on only one level of production. For an
economic analysis of this argument, see Fisher & Sciacca, supra note 8, sec. 4.
6 Krattenmaker & Salop, supra note 4, at 238, suggest that the Terminal Association was
used as a type of cartel stabilization device, or, in their terms, a "cartel ringmaster." Under
this scenario, the Terminal Association would have excluded, or threatened to exclude, a
firm from access to its facilities if that firm broke a railroad cartel agreement in another part
of the country. We did not find anything in the court record that indicated that the association ever engaged, or threatened to engage, in this type of behavior.
TERMINALRAILROADREVISITED
VI.
437
CONCLUSION
Terminal Railroad has had a curious history in the annals of antitrust.
Consistently misinterpreted, it has served as a source for misbeliefs about
the economics of vertical integration. The anticompetitive problem
identified by the court was horizontal: the combination of the Merchants
Bridge, the Eads Bridge, and the three ferry companies created a horizontal monopoly over traffic to and from St. Louis. Under current merger
standards, antitrust authorities almost certainly would seek to block the
merger or to undo it once it occurred.
The case does not support a vertical theory of antitrust harm. If a firm
has a monopoly over an input used in fixed proportions with other inputs,
then the vertically integrated firm will charge its downstream rivals the
same price it charges itself. The facts of Terminal Railroad are consistent
with the theory: association members charged nonmembers the same
price they charged themselves and denied access to no one.
Forcing the owner of an "essential facility" to provide equal access
seems to be misguided antitrust policy. In unregulated industries (with
fixed-proportion technology), there is no anticompetitive incentive to integrate. So when foreclosure does occur, efficiency considerations are a
likely motivation. The essential facilities doctrine, therefore, may discourage efficient behavior without a corresponding benefit in terms of
deterring anticompetitive conduct.
BIBLIOGRAPHY
Areeda,Phillip.AntitrustAnalysis:Problems, Tests, Cases. 2d ed. Boston: Little,
Brown, 1974.
Brennan,TimothyJ. "Why RegulatedFirms ShouldBe Kept out of Unregulated
Markets:Explainingthe Divestiture in U.S. v. AT&T."AntitrustBulletin 13
(1987):741-93.
Elzinga, Kenneth. "The AntimergerLaw: PyrrhicVictories?" Journal of Law
and Economics 12 (1969):43-78.
Fisher, Alan, and Sciacca, Richard."An Economic Analysis of VerticalMerger
EnforcementPolicy." Research in Law and Economics 1 (1984): 1-133.
Fisher, FranklinM., and McGowan,JohnJ. "On the Misuseof AccountingRates
of Returnto Infer Monopoly Profits." AmericanEconomic Review 73 (1983):
82-97.
Gerber, David. "Rethinkingthe Monopolist's Duty to Deal: A Legal and EconomicCritiqueof the Doctrineof EssentialFacilities." VirginiaLaw Review74
(1988): 1069-1113.
Katz, Michael L. "The WelfareEffects of Third-DegreePrice Discriminationin
IntermediateGood Markets."AmericanEconomicReview 77 (1987): 154-67.
Klein, Benjamin;Crawford,RobertG.; and Alchian,ArmenA. "VerticalIntegra-
438
THE JOURNAL OF LAW AND ECONOMICS
tion, Appropriable Rents, and the Contracting Process." Journal of Law and
Economics 21 (1978): 297-326.
Klein, Benjamin, and Murphy, Kevin M. "Vertical Restraints as Contract Enforcement Mechanism." Journal of Law and Economics 31 (1988): 265-98.
Krattenmaker, Thomas, and Salop, Steven C. "Anticompetitive Exclusion: Raising Rivals' Costs to Achieve Power over Price." Yale Law Journal 96 (1986):
209-93.
McGee, John. "Predatory Price Cutting: The Standard Oil (N.J.) Case." Journal
of Law and Economics 1 (1958): 137-69.
Mallela, Parthasaradhi, and Nahata, Babu. "Theory of Vertical Control with
Variable Proportions." Journal of Political Economy 88 (1980): 1009-25.
Marvel, Howard P. "Exclusive Dealing." Journal of Law and Economics 25
(1982): 1-26.
Neale, Alan D., and Goyder, D. G. The Antitrust Laws of the United States of
America. 3rd ed. New York: Cambridge University Press, 1980.
Quirmbach, Herman C. "Vertical Integration: Scale Distortions, Partial Equilibrium, and the Direction of Price Change." Quarterly Journal of Economics 101
(1986): 131-47.
Ratner, James. "Should There Be an Essential Facilities Doctrine?" University of
California, Davis, Law Review 21 (1988): 327-82.
Robinson, Joan. The Economics of Imperfect Competition, London: Macmillan,
1933.
Rogowsky, Robert A. "The Economic Effectiveness of Section 7 Relief." Antitrust Bulletin 12 (1986): 187-233.
Spengler, Joseph J. "Vertical Integration and Antitrust Policy." Journal of Political Economy 68 (1950): 347-52.
Sullivan, Lawrence A. Handbook of the Law of Antitrust. St. Paul, Minn.: West
Publishing, 1977.
Telser, Lester. "Why Should Manufacturers Want Fair Trade?" Journal of Law
and Economics 3 (1960): 86-105.
Tirole, Jean. The Theory of Industrial Organization. Cambridge, Mass.: MIT
Press, 1988.
Troy, Daniel. "Unclogging the Bottleneck: A New Essential Facilities Doctrine."
Columbia Law Review 83 (1983): 441-87.
Ulen, Thomas. "The Market For Regulation: The ICC from 1887 to 1920." American Economic Review 70 (1980): 306-10.
Warren-Boulton, Frederick R. "Vertical Control with Variable Proportions."
Journal of Political Economy 82 (1974): 783-802.
Waterston, Michael. "Vertical Integration, Variable Proportions, and Oligopoly." Economic Journal 92 (1982): 129-44.
Werden, Gregory. "The Law and Economics of the Essential Facilities Doctrine." Saint Louis University Law Review 32 (1988): 432-80.
Williamson, Oliver. "Assessing Vertical Market Restrictions: Antitrust Ramifications of the Transactions Cost Approach." University of Pennsylvania Law
Review 127 (1979): 953-93.