Booth School of Business, University of Chicago

Monopolization by "Raising Rivals' Costs": The Standard Oil Case

Author(s): Elizabeth Granitz and Benjamin Klein
Source: The Journal of Law & Economics, Vol. 39, No. 1 (Apr., 1996), pp. 1-47
Published by: University of Chicago Press for Booth School of Business, University of Chicago
and University of Chicago Law School
Stable URL: http://www.jstor.org/stable/725768
Accessed: 05-01-2016 07:12 UTC

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.

Booth School of Business, University of Chicago, University of Chicago Law School and University of Chicago
Press are collaborating with JSTOR to digitize, preserve and extend access to The Journal of Law & Economics.

http://www.jstor.org

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATIONBY "RAISINGRIVALS'
COSTS":THE STANDARD OIL CASE*
ELIZABETH GRANITZ and BENJAMIN KLEIN
Southampton College, University of
Long Island California,
University Los Angeles

ABSTRACT

Standardmonopolizedthe petroleumindustryduringthe 1870s by cartelizing

the stage of production where entry was difficult-petroleum transportation.
Standardenforced the transportationcartel by shifting its refinery shipments
among railroads to stabilize individual railroad market shares at collusively
agreed-onlevels. This method of cartel policingwas effective because Standard
possessed a dominant share of refining, a dominance made possible with the
assistance of the railroads.The railroadsfacilitatedStandard'srefineryacquisi-
tions and prevented new refinerentry by chargingdisadvantageouslyhigh rates
to non-Standardrefiners. While Standardused its dominantposition in refining
to sell refinedproductat a monopolyprice and to purchasecrude oil at a monop-
sony price, Standarddid not possess independent market power in refining.
Wheneverthe transportationcartel broke down, Standard'spricing power van-
ished.

DURING the 1870s J. D. Rockefeller established a petroleum refining

monopoly in the United States. How he accomplished this remains an
economic puzzle. In 1870 the refining industry consisted of hundreds of
small firms, and entry into the industry was easy. Rockefeller's Standard
Oil Company located in Cleveland was one of the largest refining compa-
nies but accounted for only approximately 4 percent of total U.S. refining
capacity.' In late 1871 Standard began acquiring other Cleveland refiner-
ies and in a little more than 3 months owned virtually all the refineries

* We are gratefulfor valuablecommentsfrom ArmenAlchian, Dennis Carlton, Susan

Christoffersen,HaroldDemsetz, AndrewDick, JohnHekman,GeorgeHilton, Roy Kenney,
JimLiebeler,KevinMurphy,Steve Salop,RichardSmith,JohnWiley,ananonymousreferee,
and discussantsat seminarsat UCLA, ArizonaState University,and the Universityof Chi-
cago. Valuable research assistance was provided by Laurie Carvajal,John Daly, Julie
Lawyer,andNancy Neilan.Theresearchwas supportedin partby the SouthamptonCollege-
Long IslandUniversityResearchCommitteeandby a SloanFoundationgrantto UCLA.
Testimonyof Henry M. Flagler,U.S. House of Representatives,Committeeon Manu-
factures, Reporton Investigationof Trusts(H.R. 50th Congress, Ist Sess., H.R. Rep. No.
3112, at 208 (Washington,D.C., 1888)(hereafterTrusts));and Harold F. Williamson&
ArnoldR. Daum, The AmericanPetroleumIndustry:The Age of Illumination,1859-1899,
at 344 (1959).
[Journal of Law and Economics, vol. XXXIX (April 1996)
? 1996by The Universityof Chicago.All rightsreserved. 0022-2186/96/3901-0001$01.50
1

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
2 THE JOURNAL OF LAW AND ECONOMICS

in Cleveland, or approximatelyone-quarterof U.S. refiningcapacity.2

Standardthen began makingrefineryacquisitionsin other cities and by
1879controlled more than 90 percent of U.S. refiningcapacity.3For the
following 20 years, Standardmaintaineda dominantshare of refining,in
spite of the fact that entry into refiningremainedeasy.4
John McGee in his classic article claims that Standardachieved its
dominantposition through "voluntary" merger and acquisition, buying
out refiningcompetitorsat or above marketprices.5Contraryto McGee's
analysis, however, refinershad no incentiveto sell out to Standard.Once
they recognizedthat Rockefellerwas likely to succeed, individualrefiners
would be better off holdingout and remainingoutside the Standardcon-
solidation, "free riding"on the higherindustryprice that Standardwould
create by its monopolisticrestrictionof output.Moreover,even if Rocke-
feller could induce the many refinersin the industryto consolidate into
a monopoly, it is unclear how he could maintaina dominant position
given the ease of new entry into refining.
In what follows we explain Rockefeller's success in monopolizingthe
petroleumindustryby focusing on transportationratherthan refining.In
contrast to the situation in refining, where there were numerous firms
and entry was easy, only three railroadstransportedpetroleumin 1870,
and new entry was difficult. Therefore, in principle it was possible to
establish a cartel in petroleumtransportation.Standardestablished such
a transportationcartel by collusively agreeingto stabilize individualrail-
road marketshares and by shiftingits petroleumshipmentsbetween rail-
roads to enforce the agreement.Since effective enforcementof the collu-
sive transportationagreementrequiredStandardto possess a large share
of refining,the railroadsfacilitated Standard'srefineryacquisitions and
prevented new refinerentry by setting disadvantageouslyhigh rates to
non-Standardrefiners. However, the refiningprofit earned by Standard
was only the way in which Standardreceived its portionof the petroleum
transportationcartel profits.
2 J. T. Henry, The
Early and Later History of Petroleum with Authentic Facts in Regard
to Its Development in Western Pennsylvania 315-21 (1873).
3 Testimony of H. H. Rogers and J. A. Bostwick, New York Assembly, Proceedings of
the Special Committee on Railroads, Appointed under a Resolution of the Assembly to
Investigate Alleged Abuses in the Management of Railroads Chartered by the State of New
York, at 2615, 2695-96 (8 vols. 1879-80) (hereafter Hepburn Hearings).
4 In 1899 Standard's share of U.S. refining capacity was 82 percent. By the time the
government broke up Standard in 1911 (as a consequence of Standard Oil Co. of New
Jersey v. United States, 221 U.S. 1 (1911)), Standard's share of refining capacity had fallen
to 64 percent. (See Harold F. Williamson, Ralph L. Andreano, Arnold R. Daum, & Gilbert
C. Klose, The American Petroleum Industry: The Age of Energy, 1899-1959, at 7 (1963)).
5 John S. McGee, Predatory Price Cutting: The Standard Oil (N.J.) Case, 1 J. Law &
Econ. 137 (1958).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 3

Our analysis is similarto Ida Tarbell's 1904critiqueof Rockefellerand

StandardOil, which emphasized that "Mr. Rockefeller's great purpose
had been made possible by his remarkablemanipulationof the rail-
roads."6 However, Tarbell and others who emphasized the role of the
railroads in Standard's success do not present a consistent economic
explanation of how the railroadswere "manipulated"by Rockefeller,
even in the early 1870s before Standardachieved its dominantposition
in refining.'We make economic sense of Tarbelland other commentators
by demonstratingthat Rockefeller, ratherthan "manipulating"the rail-
roads, cooperatedwith the railroadsin establishinga transportationcartel
and, in return for his cartel-policingefforts, shared in the monopoly
profits.Ouranalysis, therefore,elucidateshow a verticalrelationshipcan
facilitate the creation of monopoly power.

I. RAILROADCOMPETITION
IN THE EARLYOIL INDUSTRY
We begin with some necessary backgroundinformationon the produc-
tion and transportationof oil prior to 1870. Oil was first discovered in
northwesternPennsylvaniain 1859,in an area that became known as the
Oil Regions. For nearly 30 years essentially all U.S. oil productioncame
from this area. Figure 1 shows the location of this oil-producingarea and
the rail lines which transportedthe oil. During this period oil was used
primarilyto make kerosene, an inexpensive yet high-qualityilluminating
product.8Because the United States was the only source of kerosene,9
and because the United States was a much smallerfraction of the world
economy than today, most kerosene productionwas exported, primarily
to Europe.'0 However, before being shipped to Europe from Atlantic

6 2 Ida M. Tarbell,The Historyof the StandardOil Company67 (2 vols. 1904).

7 Williamson& Daum,supranote 1, presentsthe most completedescriptionof Rockefel-
ler's use of the railroadsto obtain a refiningmonopoly. However, this analysis does not
includea consistentanalyticaleconomicframeworkby whichone can understandRockefel-
ler's success. Morerecently, DanielYergin,The Prize:The Epic Questfor Oil, Money and
Power (1991),also accepts Tarbell'sconclusionregardingthe importanceof the railroadsin
Standard'ssuccess, but similarlyfails to identifythe crucialeconomicfactorsthat permitted
Standardto use the railroadsto achieve marketdominance.
8 Williamson& Daum, supra note 1, at 46-47, 71.
9 Very limited Russian crude oil production began in 1871 but remained insignificant for
more than a decade, accounting for less than 3 percent of world kerosene exports in 1884.
By 1889, Russian production accounted for 26 percent of world kerosene exports. See
Williamson & Daum, supra note 1, at 633.
10During 1873-75 approximatelythree-quartersof all U.S.
illuminatingoil output was
exported, with about 90 percentof exports going to Europe. As late as 1889the share of
U.S. kerosene exported was still two-thirds (Williamson & Daum, supra note 1, at 489,
633, 742).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 Railroad
Svstems R.R.
Erie
SNew YorkCentralR.R.
R.R. LakeOntari NE
.....Pennsylvania

CANADA NewYorkCentr

Buffalo Erie Canal

ke E
Erie
, & Great Westem
Atlantic ErieR.R.
LakeShore R.R. - R.R

*
• Oil'
-Regons Philadelphia&ErieR.R.
Cleveland I

* Williamsport
Allegheny River"Clve d
AlleghMiton Empire Transportati
.. -.

OHIO i
<* R.R.
Pennsylvania
Pittsburgh Harrisburg

PENNSYLVANIA "
Philadelphia
FIGURE 1.-Major railroadconnectionswith the Oil Regions, 1870.Source:HaroldF. Williamson& Arno
Industry:The Age of Illumination,1859-1899,at 299 (1959);RollandHarperMaybee, RailroadCompetit
(1940);MapInfofor Windows 1.1.3 (MapInfoCorporation,Troy, N.Y.).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 5

seaboard ports the crude oil first had to be gatheredin the Oil Regions
fields and transportedto refineries,where it was processed into kerosene.
Because refiningis a volume-losingprocess in which crude oil is dis-
tilled (boiled) at a temperaturenecessary to recover the part that is kero-
sene, and because transportationcosts were such a largefractionof total
costs, accountingfor approximately40 percent of the export price," one
might expect refineries to be located close to the oil fields. However,
refiningrequired other inputs with high transportationcosts, including
coal and chemical agents. Before the railroadsbuilt spurs into the Oil
Regions, it was difficult to transport these other inputs to the crude
source. Pittsburghwas the lowest-cost refininglocation because it had
access to coal and chemicals and was located down the Allegheny River
from the Oil Regions. Rivermen floated crude oil on barges and boats
down the Allegheny River to Pittsburgh,where it was refined and then
shipped via the PennsylvaniaRailroadto Philadelphiafor export.12
After completionin 1864of the Philadelphia& Erie Railroad,an affili-
ate of the PennsylvaniaRailroad,Philadelphiaobtaineda direct rail link
with the Oil Regions and also developed into a major refiningcenter.'3
The Pennsylvaniaalso created a direct link from the Oil Regions to New
York in 1865 by establishing the Empire TransportationCompany, an
organizationthat was given the exclusive rightto transportall petroleum
shipments(except for shipmentsoriginatingin Pittsburgh)on the Pennsyl-
vania Railroad system.14 New York's position as a refinery center and
export point was also enhanced when railroadsassociated with the Erie
Railroadand then the CentralRailroadcompleteddirect rail connections
to the Oil Regions.15

" In 1874keroseneexportprices in New York were 11.65cents per gallon, and railrates
were 4.4 cents per gallon, implyinga rail transportationcost of 38 percent of the export
price. (Crudeexport prices are from U.S. Bureauof Corporations,Reportof the Commis-
sioner of Corporationson the PetroleumIndustry, pt. 2, at 49 (2 pts. 1907) (hereafter
PetroleumIndustry);1874rail rates are fromTrusts, supra note 1, at 363.)
12 Pittsburghobtainedan all-railconnectionto the Oil Regions in 1870, when the Alle-
gheny Valley Railroad(owned by the Pennsylvania)completedits connectionbetween the
Oil Regions and Pittsburgh.See GeorgeH. Burgess& Miles C. Kennedy, CentennialHis-
tory of the PennsylvaniaRailroadCompany,1846-1946,at 173(1949).
13 See FreemanH. Hubbard,Encyclopediaof NorthAmericanRailroading:150Years of
Railroadingin the United States and Canada243 (1981);RollandHarperMaybee, Railroad
Competitionand the Oil Trade, 1855-1873,at 43 (1940).
14The Empireused this exclusive rightto providerollingstock, coordinatetraffic,and
negotiaterates with the smallrailroadsthat connectedthe Philadelphia& Erie Railroadin
Milton to New York, thereby solving the "successive monopoly" probleminherentin a
series of connectingrailroadsandfacilitatingthe growthof New Yorkrefiners.See Maybee,
supranote 13, at 49, 57; andtestimonyof A. J. Cassattin the 1879case The Commonwealth
v. The PennsylvaniaRailroadCompany;reportedin Trusts, supra note 1, at 174-75.
15 The Atlantic& GreatWesternRailroad(which was affiliatedwith the Erie Railroad)
connected the Erie system to the Oil Regions in late 1863 and the Lake Shore Railroad

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
6 THE JOURNAL OF LAW AND ECONOMICS

Anotherrefinerycenter was establishedin Cleveland,where Rockefel-

ler built his first refineryin 1863.16 Given that crude oil was located in
the Oil Regions and that kerosene was exported primarilyto Europe,
Cleveland appears to be a poor location for a refinery. Since Cleveland
is west of the Oil Regions, crude oil is moving away from the final ship-
ping point on the Atlanticcoast, therebyaddingto the total transportation
cost.7 However, Clevelandrefinerswere not disadvantagedrelativeto re-
finersin othercities becausethe two railroadsservingrefinersin Cleveland,
the Erie and the Central,had to compete with the Pennsylvaniaon a deliv-
ered-pricebasis to the East. Therefore,Clevelandrefinersobtainedcompa-
rableand at times lower railroadratesthanPittsburghand Philadelphiare-
finersin spite of Cleveland'sgreaterdistancefrom the East Coast.'8
A particularlyilluminatingepisode that illustratesthe competitionthat
existed between the three rail systems in serving their refiningcenters
occurred in 1870 when the Centralnegotiated substantiallylower rates
for year-roundcontractswith Clevelandrefiners.'"These favorabletrans-
portation rates sharply boosted Cleveland's share of shipments at the
expense of shipmentsto Pittsburghand other refiningcenters. As Table
1 indicates, Cleveland's share of crude shipmentsfluctuatedbetween 24
percent and 27 percent in the 3 years precedingthe rate cut, but after the
rate cut in 1870 the share rose to more than 35 percent.20In May 1871,

(which was affiliatedwith the New York Central)establishedan effective link to the Oil
Regions in 1870. See Maybee, supra note 13, at 16-17, 223.
'6 1 Allan Nevins, Study in Power:John D. Rockefeller,Industrialistand Philanthropist
41 (2 vols. 1953).
17 Some very small fractionof worldwidekerosene demandwas in the western part of
the United States for which Clevelandwas a preferredrefininglocation.
18 For example, in 1867the rate to ship refinedproductfrom Pittsburghto Philadelphia,
355 miles along the PennsylvaniaRailroad,was $1.70 per barrel.At the same time the rate
from Clevelandto New York, 629 miles, was only $1.53 per barrel;see Maybee, supra
note 13, at 235. Clevelandalso had access in the summerto a water route on shipments
east, via the Great Lakes, the Erie Canal, and then the Hudson River. This water route,
however, could not completelyavoid the railroadsbecausethe crudefirsthad to be shipped
from the Oil Regions to Cleveland(or to anothercity on the GreatLakes) by rail.
19The contractsstipulatedthe minimumquantitythat refinershad to ship over the year
by railto receive the discount.For example,Standard'scontractwiththe Central'saffiliated
Lake Shore Railroad granted Standard, in exchange for guaranteed shipments, a rate of
only $1.30 on shipments of refined products to New York, while the open rate ("list price")
at the time was $2.00. According to James Devereux, Vice President of the Lake Shore
Railroad,"[T]hiswas the turningpoint which securedto Clevelanda considerableportion
of the export traffic."See Devereux Affidavit,1 Tarbell,supra note 6, at 278-79.
20 The decline in the shareof cruderefinedin other centers, such as Pittsburghand New
York, did not translateinto an absolutedecline in the quantityof barrelsrefinedat these
locations because over this period the total volume of crude shipmentswas increasing
dramatically.For example,from 1869to 1870the total quantityof crude shippedincreased
by more than 30 percent(Trusts,supra note 1, at 115).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 7

TABLE 1
PERCENTAGE
OF CRUDE SHIPPEDTO MAJORREFININGCENTERS, 1865-71

REFININGCENTER NEW YORK, OIL

REGIONS, AND
YEAR Cleveland Pittsburgh Philadelphia OTHERCENTERS
1867 24.0 25.1 33.6 17.4
1868 27.5 30.5 30.6 11.5
1869 26.4 24.1 24.7 24.9
1870 35.8 18.6 26.4 19.3
1871 26.5 20.2 23.5 29.9

SOURCE.-Cleveland: Rolland Harper Maybee, Railroad Competition and the Oil

Trade, 1855-1873, at 224, from the Annual Statement of the Trade, Commerce, and
Manufacturesof the City of Clevelandfor the years 1865, 1866, 1867, 1868, 1869,
1870. Pittsburgh: Maybee, supra, from the Pittsburgh Commercial, February 22,
1872; also in J. T. Henry, The Early and Later History of Petroleum with Authentic
Facts in Regardto Its Developmentin WesternPennsylvania309(1873).Philadelphia:
Maybee,supra, at 308. (Dataappearunderthe heading"receiptsof [p]etroleumby
railroad, at Philadelphia." This definition may not be limited to crude oil receipts;
refinedoil shippedfrom Pittsburghto Philadelphiamay have also been included.)
New York, Oil Regions, and other centers: these results were calculatedby sub-
tractingthe Pittsburgh,Cleveland,and Philadelphiadata fromtotal shipments(U.S.
House of Representatives, Reports on Investigation of Trusts (H.R. 50th Congress,
Ist Sess., H.R. Rep. No. 3112, at 115 (Washington, D.C., 1888)). They consist primar-
ily of shipments to refineries in New York and the Oil Regions, but also include
other smaller refining centers during the period such as Boston, Baltimore, Erie, and
Buffalo.

in response to these relatively low rates on Cleveland shipments, the

Empire TransportationCompany cut the rate on crude oil shipped to
New York from the Oil Regions from $1.90 to $1.50.21This change in
relative rates caused the New York (and Oil Regions) refiners' share of
the marketto increase from 19 percent to 30 percent, while Cleveland's
share fell from 36 percent to 26 percent, back to where it was in 1869
(before the initial Cleveland rate cuts were made). It is clear that the
three railroad systems were competing with one another and that the
large year-to-yearvariationsin the fraction of crude shipments moving
to alternativerefiningcenters duringthis period can be largely explained
by changes in relative rail rates between the alternativecenters.
The competitive pressures on the three railroadsto reduce rates on
petroleumshipmentswere substantial.Because the overwhelmingmajor-
ity of oil was shippedto the East Coast for export to Europe, the particu-
lar destinationcity on the East Coast was not an importantconsideration
for oil producers and refiners. Therefore, the three railroadsinvolved
in the transportationof oil were close substitutes for one another. The
competitionfaced by these three railroadson petroleumshipments was
21 Maybee, supra note 13, at 266.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
8 THEJOURNALOF LAWAND ECONOMICS

no differentthan the competitionrailroadsfaced at the time for shipping

any product on a route served by multiplerail lines. Duringthis period,
before the establishmentof the InterstateCommerceCommission(ICC),
railroadpricingwas largelyunregulated.22 As a result, most railroadshad
local monopoly power on short routes but faced significantcompetition
on longer routes. The railroads'high sunk costs (notably, investments in
the road bed) and relativelylow variablecosts meantthat railroadsserv-
ing the same route engaged in rate wars. This often resulted in very low
rates (that is, prices substantiallybelow a railroad'saverage costs) and
in long-haul versus short-haulrate discrimination,all part of what was
referredto at the time as "the railroadproblem."23
In the face of "ruinous" rate competition, the railroads repeatedly
establishedcartel or poolingarrangements.24However, cartelrate setting
invariablylead to cheating, with railroadsgiving secret rebates to large
shippers or intentionallymisclassifyingfreight, which eventually lead to
rate wars and rates being driven down to variablecost. As HerbertHo-
venkamp has accurately characterized, "In no other industry have at-
tempts at both legal and illegal cartelizationbeen so persistent, wide-
spread, systematic, or ultimatelydoomed to failure."25
To sum up, at the time Rockefellerbeganhis acquisitionsin 1871there
were three majorrefininglocations-Cleveland, Pittsburgh,and Philadel-
phia, each accountingfor about a quarterof U.S. refiningcapacity. New
York, the Oil Regions, and some other less significantrefininglocations
together accounted for the remainingquarterof capacity. At this time
three majorrail lines (or combinationsof rail lines) tied the Oil Regions
to refineriesand to the seacoast. Two railroads,the Erie and the Central,

22 Before passage of the InterstateCommerceAct in 1887, some States regulatedrail

rates, but these regulationswere neitheruniformnor very effective. See GeorgeW. Hilton,
The Consistencyof the InterstateCommerceAct, 9 J. Law & Econ. 87 (1966).
23 Thereis an extensive contemporary literaturethatdocuments"the railroadproblem."
See, for example,CharlesFrancisAdams,Jr., Railroads:TheirOriginandProblems(1878);
Lloyd Bryce, The Railway Problem:I. The Legislative Solution, 164 N. Am. Rev. 327
(1897);A. B. Stickney,The RailroadProblem(1891);JamesJ. Wait,The RailwayProblem:
II. A MercantileView, 164N. Am. Rev. 338 (1897).
24 Before passage of the ShermanAntitrustAct in 1890, explicit collusive agreements
were not per se illegal.Whilethe Englishcommon-lawtradition,adoptedby manyAmerican
state courts, consideredcollusive agreementstorts if they were not ancillaryto some other
business purpose, the Englishcourts exemptedcollusive behaviorby railwaysbecause of
their alleged naturalmonopolystatus. Americancourts never definitivelydecidedwhether
they followed the Englishcourts in that exemption.This left the railroadcartels somewhat
in legal limbo. As a result, the railroadsestablished explicit collusive agreements, but
generallydid not attemptto enforce these agreementsin court. See Hilton, supra note 22.
25 HerbertHovenkamp,RegulatoryConflictin the GildedAge: Federalismand the Rail-
road Problem,97 Yale L. J. 1017, 1039(1988).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 9

tied the Oil Regions to Clevelandand to New York, while one railroad,
the Pennsylvania,tied the Oil Regions to Pittsburgh,to Philadelphia,and
to New York. Competition existed between all three rail lines, and
changes in crude flows from the Oil Regions to the alternativerefining
centers were determinedin partby differentialtransportationratesamong
the three competingroutes.

II. THE SOUTHIMPROVEMENT COMPANYCONSPIRACY:ROCKEFELLER'S

OF
ACQUISITION CLEVELAND
REFINERIES
(1871-72)
The three railroads,aware of their repeatedfailuresto collude on rates
and the existence of "ruinous"rate competition,decided in 1871to enlist
the cooperation of several refinersto police a collusive rate agreement.
The railroadsestablished a corporationcalled the South Improvement
Companyand selected the largest refinersin each of the refiningcenters
of Pittsburgh,Philadelphia,Cleveland, and New York as joint owners of
the company.26Standardwas chosen as the South ImprovementCom-
pany refiner in Cleveland.27The cooperatingrefinersin each city were
referredto as "eveners" because the South ImprovementCompanycon-
tract with the railroadsrequired the refiners to adjust their shipments
between the railroadsin order to maintaineach railroad'scontractually
specified marketshare of petroleumshipments.The contractuallyspeci-
fied marketshares were 45 percent of petroleumshipmentsto the Penn-
sylvania Railroadand 27.5 percent each to the Erie and Central.28This
eliminated the incentive for railroadsto cheat on the jointly set rates
because any increase in shipmentsgained by a railroadby lowering its
rates would be offset by an evening refinerdecreasingits own shipments
on the railroad.
In return for this evening service, the member refiners of the South
ImprovementCompanyreceived a "rebate" on their own shipmentsand

26 The idea for the companyis generallyattributedto ThomasScott of the Pennsylvania

Railroad.See 1 Nevins, supra note 16, at 103.
27
The chosen refinerin New York, J. A. Bostwick, mergedwith Standardon January
1, 1872, 1 day before the South ImprovementCompanystock was subscribedto by the
shareholders.When combined with Bostwick, Standardheld a 45 percent share of the
SouthImprovementCompany.The SouthImprovementCompanyrefinerin Pittsburghwas
Lockhart,Frew & Co., and in Philadelphiait was Warden,Frew & Co. In addition,Peter
H. Watson,the Lake ShoreRailroadexecutivewho was the presidentof the SouthImprove-
ment Company,held 5 percent of the Company'sstock and on January1, 1872, also ac-
quired500 sharesof Standardstock. See testimonyof W. G. Warden,in PetroleumProduc-
ers' Union, A History of the Rise and Fall of the South ImprovementCompany30-31
(1872) (hereafterPetroleumProducers'Union); 1 Nevins, supra note 16, at 135.
28 The contracts were signed on January 18, 1872, and are reproducedin Petroleum
Producers'Union, supranote 27, at 97-120.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
10 THEJOURNALOF LAWAND ECONOMICS

a "drawback" on their competitors' shipments. The rebate granted to

South ImprovementCompanyrefinerswas paid at the time the member
refiner shipped, so that the South ImprovementCompany refiner only
paid the railroada rate net of the rebate. Competingnonmemberrefiners
paid the full gross rate on shipment, and the railroadspaid a drawback
on these shipments at the end of the month to the South Improvement
Company. The South ImprovementCompanywas to inspect the books
of the three railroads, collect all the shipment data, receive the draw-
backs, and distributethe drawbackrevenue to its shareholders.29Thus,
a refineroutside the South ImprovementCompanypaid its competitors
whenever it shippedoil. The South ImprovementCompanycontractcre-
ated a profitablelong-termprice structurefor the participatingrailroads,
with the chosen refiners (who made the profits possible) sharing the
profitswith the railroads.30
Table 2 indicatesthat the rates proposedby the railroadsin their agree-
ment with the South ImprovementCompanyrepresenteda huge increase
over preexistingrates. For example, on refinedshipmentsfromCleveland
to New York, rebates to all refinersother than Standardwould be elimi-
nated, and rail rates would increase to $2.00, the previous "open" or list
price. Standard,even with its $.50 rebate, would pay a net transportation
rate of $1.50, or 15 percent more than the $1.30 rate it previously paid.
Since the rebate and the drawbackwere the same, $1.50 was the net rate
(net of rebates or drawbacks)the railroadsexpected to collect on all
similar shipments. Hence the rates received by the railroadswould in-
crease $.20, or 15 percent, from normalrebatedlevels and substantially
more on averageif the South ImprovementCompanycontracteliminated
the periodic rate wars. The evening refiners,while they paid these higher
rates on their own rail shipments,were also substantiallybetter off. They
collected a $.50 per-barreldrawbackon theircompetitors'shipmentsand
received a higherprice on their own final sales.
To analyze the effect of the South ImprovementCompanyagreement

29 South ImprovementCompanyContract, in id. at 97-98. The rebate and drawback

rates were set at the same level so that the railroadshad no incentiveto deviate from the
agreed-onrail rates. For example, if the rebatewas greaterthan the drawback,a railroad
would have an incentive to lower rates because, althoughevening shipments would be
moved away from the railroadto keep its marketshareunchanged,the railroadwould end
up with a largerfractionof noneveningrefinershipmentsand hence highernet revenues.
30A similarevening arrangement,facilitatedby the meat-packers,was used for cattle
shipmentsto Chicago.(See testimonyof R. C. Vilas, HepburnHearings,supra note 3, at
397-98; also see Gilbert Holland Montague,The Rise and Progressof the StandardOil
Company 17 n.2 (1903). The railroads'attemptsto establish similararrangementsacross
multipleproductsby establishingclearinghousesgenerallyfailed for lack of an effective
enforcementmechanism.See Adams, supra note 23, at 170-79.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 TABLE 2
RAILROAD
PETROLEUM RATES, 1870-72 (Dollars per Barrel)

SOUTH IMPROVEMENT
1870 COMPANY, 1872

Evening
Open Standard's Gross Refiners' Net O
Rates Rates* Rate Rebate Rate A
Crude:
From the Oil Regions to:
Cleveland . . . .35 .80 .40 .40
Pittsburgh ... ... .80 .40 .40
New York ... ... 2.56 1.06 1.50
Philadelphia ... ... 2.41 1.06 1.35
Refined:
From Cleveland to New York 2.00 1.30 2.00 .50 1.50
From Pittsburgh to New York ... ... 2.00 .50 1.50
From Oil Regions to New York ... 2.92 1.32 1.60

SOURCE.--Petroleum railroad rates for 1870 are from Rolland Harper Maybee, Railroad Competition and the Oil Tr
Improvement Company contract rates for 1872 are from Petroleum Producers' Union, A History of the Rise and Fall of the
(1872); new April 1, 1872, open rates were established on termination of the South Improvement Company contract, by agr
Union on March 25, 1872, from U.S. Bureau of Corporations, Report of the Commissioner of Corporations on the Petro
1872 Standard rates are from the testimony of Henry M. Flagler, reprinted in I Ida M. Tarbell, The History of the Standa
* Other refiners also received substantial discounts from open rates during this period.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
12 THE JOURNAL OF LAW AND ECONOMICS

+MCt
fFCc+r
Sc+r + MCt
MFCr
Nfc+r ..'"
P . ....... .. . ..... Sc
M, ........" ?r

... .....
.....
... ....

A
........ .... .
..... f I
', 00"0
------ - ----
ED -
.~

D .. 11Q... 0--?---- Q
..... ...... ---- I~ m c

c ---------------
--o--

FIGURE
2.-The South ImprovementCompanyconspiracy

we assume for simplicitythat there are three inputsor "stages of produc-

tion" in the petroleumindustry-crude production,transportation,and
refining. We further assume that there is no substitutionamong these
three inputs and that each input is used solely to produce petroleum.
Therefore,a monopoliston any one inputcan be thoughtof as purchasing
the other two inputs as a monopsonistand selling the final product as a
monopolist.
Figure 2 illustrates the situationfor the case where transportationis
monopolized.The long-runsupplyconditionsfor the firsttwo competitive
stages, crude production(Sc) and refining(S,), are added vertically to
form Sc,,r the supply of refinedkerosene facing firmsat the "last" stage
of production,railroadtransportation.Railroadsare assumedto purchase
refinedkerosene from refinersand then to sell the productin the export
market.31If the railroadtransportationindustrywere perfectly competi-
tive, the marginalcost of railtransportation,MC,, could be addedto Sc+,

31 The assumptionthatthe railroadsare the "last" step in the productionprocess is made

for expositionalconvenience.Givenperfectcomplementarityof inputs,this assumptionhas
no effect on the results of the analysis.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 13

to determinethe industrysupplycurve for kerosene, which is represented

in Figure 2 by the dotted line Sc+, + MC,. Under these conditions the
competitive price of refined kerosene would be PC and the output of
kerosene Q, with the crude suppliersreceiving a crude price of A, the
refiners receiving the difference between B and A as a refinery margin,
and the railroadsreceivingthe differencebetweenPCandB as the railroad
rate.32
If, on the contrary, there were a monopoly railroadpurchasingkero-
sene from refinersmonopsonisticallyand selling the refinedkerosene mo-
nopolisticallyin the export market,it would maximize its profits by first
determiningthe marginalfactor cost of Sc+,, representedin Figure 2 as
MFCc+r,, and then addingits marginalcosts of producingtransportation
services, MCt, to obtain its marginalcost of purchasing, shipping and
selling additionalbarrelsof refinedproduct,MFCc+r+ MCt. The monop-
olist railroadwould set the final refinedproductprice where MFCc+r +
MCt equals the marginalrevenue of the industrydemandcurve. This is
illustrated in Figure 2 at the point Pm, Qm.33
A railroadmonopoly reduces the returnsto both crude suppliers and
refiners,as quantityfalls and these other inputsuppliersmove down their
supply curves. Since crude supply is likely to have been highly inelastic,
at least over the short term, a monopolistrailroadcan collect significant
rents from crude suppliers. This is illustratedin Figure 2 by the crude
price decreasingfrom A to C. Refiningcan be expected to have a much
more elastic supply, as representedin Figure 2 by the fact that Sc+r is
not much steeper than and the per-barrelrefinery returns only de-
Sc,
crease from (B - A) to (D - C). The difference between Pm and D is

32 The
equilibriumdescribed here, where the competitive railroadrate is equal to the
marginalcost of providingrailtransportationservices, is not a long-runequilibriumbecause
such a rate is less than the averagecost of providingtransportationservices and hence the
railroadswouldreceive no returnon the capitalthey investedin the transportationnetwork.
However, for expositionalsimplicitywe assumeat this pointthatinvestmentsin the railroad
networkare preexistingand nonsalvageableand that this is the competitiveequilibrium.
33 A monopolist vertically integratedinto all three stages would appear to be able to
eliminatethe successive monopoly(or monopsony-monopoly)element of price determina-
tion and therebysell moreunitsat a lower finalprice. However, if the monopolistvertically
integratesinto crude productionby purchasingcrude fields, the monopsonydistortionre-
mains. The distortionis merelyshiftedfromthe purchaseof increasedquantitiesof current
barrels of oil to the purchaseof increased quantitiesof currentand future expected oil
supply-or from flow prices to capitalprices. In particular,as the monopsonistdecides to
buy up more crude producingproperties,it will have to take account of the fact that it will
bid up the prices of all properties.This may explain why Rockefellerowned only a trivial
amount of crude producingpropertiesin the Oil Regions. For example, as late as 1888,
when total U.S. productionwas 76,000 barrelsper day, Standard'scrude productionwas
only 200 barrelsper day (RalphW. Hidy & MurielE. Hidy, Pioneeringin Big Business,
1882-1911, at 175-76 (1955)).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
14 THE JOURNAL OF LAW AND ECONOMICS

the per-barrelmonopoly railroadrate, which is substantiallyabove the

marginalcosts of providingrailroadtransportation,E - D, implying a
per-barrelmarginalprofitof Pm - E. This per-barrelmarginalprofitcon-
sists of two components-an increase in the final productprice from
Pc
to Pm and a reduction in the final product cost from Pc to E, primarily
due to lower crudeprices. The SouthImprovementCompanycontractual
arrangementintendedto share this monopoly-monopsonymarginalprofit
among the evening refinersand the railroads.
Whileincreasingrailroadrates decreasedthe crudeprices paid by inde-
pendent refiners and increased the refinedprices at which independent
refiners sold their output, the open rail rates set by the South Improve-
ment Company contract pushed independent refiners into a negative
profitposition.34The rates were set above the monopoly level, Pm - D,
thereby "squeezing" independentrefinersand forcingthem to sell out to
the evening refiners.Standard,in particular,aggressivelytook advantage
of this to acquirecompetingrefinersat low prices. In the 3-monthperiod
between the end of December 1871 and April 1, 1872, after the South
Improvement Company was established and formal agreement was
reached with the railroadson rates, but before the South Improvement
Companycontracts were scheduled to go into effect, Standardacquired
all the majorrefineriesin Cleveland, thereby obtainingcontrol of about
25 percent of the nation's refiningcapacity.35
The South Improvement Company never went into effect because
crude producers blocked its implementation.Two days after an officer
of the Lake Shore Railroadprematurelyannouncedon February25, 1872,
that the rate on crude shipmentsto New York had been increased from
$.87 to $2.14 a barrel,3,000 men gatheredin the Titusville OperaHouse
in the heart of the Oil Regions and formedan organizationof crude pro-
ducers called the PetroleumProducers'Union. The PetroleumProducers'
Union imposed an embargoon crude shipmentsto the South Improve-
ment Company refiners, with crude producers who sold to South Im-
provementCompanyrefinersrunningthe risk of personalinjuryand dam-
age to their property. The embargo, while not complete, was very
effective. These actions, known at the time as the Oil War, led the rail-
roads on March 25, 1872, to cancel their contracts with the South Im-

34 See note 53 infra for the calculation of actual independent refiner profits under the
similar differential rate situation that existed in 1878.
35 Standard began its purchases in Cleveland with the Clark, Payne & Co. refinery. This
acquisition was amicable and involved an "extra" payment for goodwill, with Payne joining
Standard's management. However, as we discuss below, many of the later acquisitions
were much less amicable. See 1 Nevins, supra note 16, at 134-36; Henry, supra note 2, at
315-21.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 15

provement Company and on April 1, 1872, to agree on a new schedule

of reduced rail rates with the PetroleumProducers'Union. In addition,
the PetroleumProducers'Union broughtpublic pressure to bear on the
railroads,leadingthe Pennsylvanialegislatureon April2, 1872,to revoke
the South ImprovementCompanycharter.36
Because of the rapidcollapse of the South Improvementarrangement
McGee claims that it is unlikelythat Standardused it to acquireCleveland
refineries.37However, the timingevidence, namely, that all the Cleveland
acquisitionsoccurredin essentially a 3-monthperiod after the South Im-
provementCompanywas establishedbut before the Companywas forced
to disband, suggests that Rockefellerdid use the threatof his prospective
transportationcost advantage as a member of the South Improvement
Companyto induce his Clevelandrivals to sell out.
Moreover, in additionto the timingof the acquisitions,there is consid-
erable testimonial evidence that Standardused the South Improvement
Companyto induce Clevelandrefinersto sell and that many of these sales
were at distress prices. Several Clevelandrefinerstestified that they sold
their plants to Standardat prices below what they believed was market
value because they faced much higherrailroadrates from the South Im-
provement Company agreement and railroadexecutives or South Im-
provementCompanyofficialsinformedthem that they would always face
such disadvantageousrates. A vivid exampleis testimonyof Mr. Basling-
ton of the firmof Hanna, Baslington,and Company,a refinerpurchased
by Standardon February22, 1872. Mr. Baslingtonstates that
afterhavinghad an interviewboth with Mr. Watson,who was presidentof a
company called "The South ImprovementCompany," and Mr. Devereux, who
was generalmanagerof the Lake ShoreRoad, he becamesatisfiedthat no arrange-
mentwhatevercouldbe effectedthroughwhichtransportation
couldat leastbe
obtainedon the LakeShoreRoadthatwouldenabletheirfirmto competewith
the StandardOil Company, the works of said Hanna, Baslington& Company,
beingso situatedthatthey couldonlyobtaintheircrudethroughthe line of the
Lake Shore Road. .... [U]nderthese circumstancesthey sold their works to the
Standard
OilCompany,whichwereonthedayof thesaleworthatleast$100,000,
for $45,000because that was all they could obtainfor them, and works too which
in cashcost themnotless than$76,000,andwhichwitha faircompetition
would
havepaidthemanincomeof notlessthan30percent.perannumoninvestment.38
36 Williamson& Daum,supranote 1, at 350-51; PetroleumProducers'Union, supranote
27, at 5-17, 26-27.
37 McGee, supra note 5, at 145, n.19.
38 Affidavitof GeorgeO. Baslington,in the case of the Standard Oil Companyv. William
C. Scofield et al., in the Court of CommonPleas, CuyahogaCounty, Ohio (reprintedin
1 Tarbell,supra note 6, app. 7, at 290-91). Similartestimonialevidence of distress sales at
the urgingof Watson is provided, for example, by Mr. Hewitt, a partnerin Alexander,
Scofield & Company, a Clevelandrefinerpurchasedby Standardon February21, 1872

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
16 THE JOURNAL OF LAW AND ECONOMICS

Contraryto a normalmergerfor monopoly case, independentrefiners

did not have an incentive to hold out for a higherpurchaseprice. In the
normal case, once independentrefiners anticipated that Standard was
likely to succeed in creatingmarketpower in refining,they would expect
a higher refined product price from Standard'sexpected restriction of
output. Therefore, independentrefinersthat remainedoutside the Stan-
dard group would face the likelihood of a price umbrellaunder which
they could expandtheir outputand increasetheirprofit.However, in this
case Standardand the other evening refinerswere not creatinga refining
monopoly, but were policing a railroadcartel. Because rail rates and
not refinerymarginswere anticipatedto rise, outside refinersfaced the
possibility not of a price umbrella,but of a price squeeze. If Standard
and the other evening refinerssucceeded in policinga transportationcar-
tel, the refinedprice facing outside refinerswould not increase as much
as their transportationcosts. Although the independent refiners knew
there was some probabilitythe collusive railroadpricing scheme would
not survive, this probabilityonly decreased the discounted value of the
expected squeeze; it did not eliminatethe possibilityof a squeeze. There-
fore, independentrefinerswould willinglysell out to Standardat distress
prices that reflectedthe probabilitythe collusive scheme would succeed."
Railroad executives facilitated Standard'sacquisitions in Cleveland,
emphasizing the future threat to independent refiners of substantially
higher, disadvantageousrail rates, in orderto increase the evening refin-
ers' market share and thereby stabilize the transportationcartel. While
the acquisitionsincreased Standard'sbuyingpower in the transportation
market, without the acquisitionsand a substantialexpansion of the eve-
ning refiners' share the collusive agreementwould not have survived.
Because the services of the three railroadswere essentiallyperfect substi-
tutes and each railroad'smarginalcost of providingadditionaltransporta-
tion services was extremelylow, the short-runprofitgains from cheating

(Hepburn Hearings, supra note 3, at 2528-30), by Mr. Alexander at an 1872 hearing (Petro-
leum Producers' Union, supra note 27, at 50-52), and at an 1876 congressional hearing
(1 Tarbell, supra note 6, at 65). While the self-serving financial details of this testimony
must be accepted with some skepticism, one cannot ignore the fact that a large number of
refiners told a similar story and, in particular, emphasized the role of Watson.
39 Standard had the additional incentive to make acquisitions because doing so transferred
profits from the other South Improvement Company refiners to itself. By making acquisi-
tions Standard collected "up front" the total future drawback payments the independent
refiner expected to make to the South Improvement Company. Absent the acquisition,
Standard would only have collected a share of these drawback payments, with the remain-
der being collected by its South Improvement Company partners. The eveners in the other
cities were similarly motivated to make acquisitions. For example, Lockhart & Frew, the
evener in Pittsburgh, controlled seven plants accounting for more than half of Pittsburgh's
refining capacity by 1874. See 1 Nevins, supra note 16, at 209-10.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 17

on the collusive arrangementwere extremelyhigh. The cheatingthat had

occurred under previous railroadcollusive agreementsinvolved not just
secret discounts on marginalsales, which evening refinerswho controlled
15 percent of industryshipmentscould have smoothedout, but also wide-
spread and massive discounts from agreed-onrates offered to essentially
all shippers. Hence, for the evening refiners effectively to police the
South ImprovementCompany agreement it was necessary to increase
their marketshare substantially.The railroadscooperatedwith Standard
because they knew that without Standard'sacquisitionsthey faced a re-
turn to ruinousrate wars, whereas with Standard'sacquisitionsthe com-
petitive pressureson theircollusive agreementwould be greatlyreduced.

III. ROCKEFELLER'SCONSOLIDATIONOF REFINERY OWNERSHIP (1872-79)

After the South ImprovementCompanywas disbanded, the railroads

and the PetroleumProducers' Union agreed on new open rail rates (or
"list prices") on April 1, 1872.These rates, an illustrativesubset of which
are presentedin Table 2, were very close to the net rates proposed in the
South ImprovementCompanycontract.The railroadsand the Producers'
Union, perhaps disingenuously,included a statementwith the new rate
announcementproclaimingthat all rates "after this date shall be upon a
basis of perfect equality to all shippers, producersand refiners,and that
no rebates, drawbacks,or other arrangementsof any character,shall be
made or allowed that will give any party the slightest difference in rates
or discriminationof any characterwhatever."40 Withoutrebates the rail-
roads would be as well off as under the South ImprovementCompany
contract, while crude suppliersand independentrefinerswould be much
better off. Only the evening refinerswould be disadvantagedrelative to
their position under the South ImprovementCompanycontract.
Past experience would suggest that, withouta policing mechanism,the
new open rates were unlikely to stick and that rebates from these rates
soon would be offered. As expected, within2 weeks Standardnegotiated
a preferentiallower rate from the New York Central. In exchange for
guaranteedshipments, Standardnegotiateda rate of $1.25 in the summer
months and $1.40 in the winter months (or an average rate of $1.31) for
refined shipments from Cleveland, when the open rate was $1.50.41 In
1874 Standardalso used share guaranteesto negotiatea preferentialrate

40
Petroleum Producers' Union, supra note 27, at 27.
41 Flagler Testimony, reprinted in 1 Tarbell, supra note 6, at 332-33. The average rate
is calculated by weighting the two rates by monthly shipments over the 1872-73 period.
See The Derrick's Hand-book of Petroleum: A Complete Chronological and Statistical
Review of Petroleum Developments from 1859 to 1898, at 807 (1898) (hereafter Derrick's).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
18 THE JOURNAL OF LAW AND ECONOMICS

agreementwith the Erie Railroad.42 The surprisingthing is not that Stan-

dard negotiatedpreferentialrates, but that these preferentialrates were
comparableto, and actuallyhigherin real terms, than the rates Standard
had negotiatedin 1870,before it acquiredits dominantposition in Cleve-
land and its greatly enhancedbargainingposition with the Erie and Cen-
tral.43This comparisonclearly suggests that Standarddid not fully exer-
cise its increased buying power when negotiatingrail rates.
Standard'sdecision to "pull its punches" and pay relatively high rail
rates to the Central and Erie can be understood in the context of the
failed South ImprovementCompany.Standardcould have used its much
larger relative size in Cleveland to negotiate more favorable rates with
the Erie and the Centralrailroads.However, expected profitsfrom such
reduced railroadrates were much less than what Standardwould collect
if it establishedand policed a railroadcartel similarto the South Improve-
ment Company arrangement.While reduced railroadrates would give
Standardsome temporaryadvantage,competitionby the Erie and Central
Railroads with the PennsylvaniaRailroadwould lead the Pennsylvania
to grant similarrate rebates to refinersin Pittsburgh,Philadelphia,and
New York. As a result, most of the reductionin Standard'srailroadrates
would show up in lower refinedkerosene prices and highercrude prices.
Therefore,ratherthan using its dominantposition in Clevelandto negoti-
ate favorablerates for its shipmentsfrom the Erie and Centralrailroads,
the evidence indicates that Standarddecided to attemptto duplicate the
South ImprovementCompanyarrangementand establish itself as a po-
licer of monopoly railroadrates.
For Standardto be a railroadrate policer, it needed a way to move
shipments away from the PennsylvaniaRailroad. Without this ability,
the Pennsylvaniacould undercutthe relatively high rail rates Standard
establishedon the Centraland Erie and shift Oil Regions crude shipments
away from Standard'srefineriesin Clevelandto the independentrefiner-
ies in Pittsburgh,Philadelphia,New York and the Oil Regions. To gain
leverage over the Pennsylvania,Standardsecretly merged in 1874 with
its originalpartnersin the South ImprovementCompany,who were then
the largest refinersin Pittsburghand Philadelphia."As opposed to most
42 The 1874 agreement between the Erie and Standard is in
Hepburn Hearings, supra
note 3, at 3398-3402.
43 A nominal rail rate of $1.31 in 1872 was equivalent to a real rail rate of $1.36 in 1870
dollars because the United States was experiencing price deflation during the 1870s. (The
U.S. price level used for this calculation is taken from Milton Friedman & Anna J.
Schwartz, Monetary Trends in the United States and the United Kingdom: Their Relation
to Income, Prices, and Interest Rates, 1865-1975, at 122 (1982), table 4.8.)
4 1 Nevins, supra
note 16, at 209-210. Standard had already merged with its New York
refinery partner in the South Improvement Company (note 27 supra).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 19

of the acquisitionsin Cleveland,these mergerswere amicableon all sides.

The principalsof the mergedorganizationscontinuedrunningtheir refin-
eries and became shareholdersand officers in the Standardorganization.
Standardnow owned fully 40 percent of U.S. refining.45 More important,
Standardcould now pressurethe Pennsylvaniawith the threatof moving
its crude shipmentsaway from its newly acquiredrefineriesin Pittsburgh
and Philadelphiato its Clevelandrefineries.46This action would be some-
what costly to Standard(given risingmarginalcost of refiningoperations),
but it would be much more costly to the PennsylvaniaRailroad.Standard
now had the ability to pressure all three railroads.
Once Standardcould shift its oil shipmentsamong all the rail lines, it
acquiredthe ability to set up a collusive rail rate policing mechanismas
originallyenvisioned in the South ImprovementCompanyagreement.In
1875 Standardnegotiated a 10 percent rebate from the Pennsylvania.47
This equaled the rebate Standardreceived at the time from the Erie, the
Central, and a new entrantinto petroleum shipping, the B&O Railroad
(which we discuss below). In returnfor preferentialrates, Standarden-
forced the shipmentshares agreed to by the railroads,dividing its ship-
ments among the railroadsso that the Pennsylvania"got 52 percent [of
the traffic]. . . . The Baltimoreand Ohio got 9 percent and the remainder
was divided equally between the New York Centraland the Erie."48
The evidence is clear that Standard'spreferentialrates were set at a
relatively high level. Although the railroadsguaranteedStandardlower
rates than any of its rivals, these rates were significantlyabove the mar-
ginal cost of providingrail transportationand above the rates negotiated
by Standardin 1870duringthe "ruinouscompetition"period. Moreover,
the rates were significantlyabove the rates charged for shipping other
commodities. Mr. Blanchardof the Erie Railroad,in describingthe rates
paid by Standardover the 1874-78period, statedthat "continuouslyfrom
the first contract made by the StandardCompanywith us on the 17thof
April, 1874,this business had averaged2 to 5 times as much per hundred
pounds as the other throughfreight that we have carried, per ton per

45 Calculation based on data in Henry, supra note 2, at 315-21.

46 During this period Standard utilized only about half of its capacity in Cleveland. See
1 Nevins, supra note 16, at 192. Evidence that this threat to the Pennsylvania was real can
be found in the behavior of Standard during the Empire Transportation Company rate war
of 1876-77, infra at Section VA.
47 Blanchard testimony, Hepburn Hearings, supra note 3, at 3450-51.
48 Testimony of A. J. Cassatt, Vice President of the Pennsylvania Railroad, in the case
of The Commonwealth v. The Pennsylvania Railroad Company, 1879, reprinted in Trusts,
supra note 1, at 199. Cassatt states (at 184) that "[t]he Standard Oil Company never had
anything to say as to what the division should be between the lines."

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
20 THEJOURNALOF LAWAND ECONOMICS

mile. ... [I]t is clearly the most profitablebusiness that the Erie Railway
has done of all its throughfreighteastward.""49
Table 3 presentsconfirmingevidence on railratesfor variouscommodi-
ties duringthis period. As we can see, the petroleum rates set in 1874
were substantiallyhigher than the contemporaneousrates for coal and
grain.5oMoreover, the increase in petroleumrates that occurredin 1874
was contrary to the downward movement in grain rates that was oc-
curringover the same period, resultingin a dramaticincrease in petro-
leum rates relativeto grainrates. (The laterdecrease in relativepetroleum
rates that occurs after 1879is discussed infra.)
Although petroleum rail rates were relatively high during 1874-79,
Standard'stransportationrates were substantiallylower than their refin-
ing competitors' rates. Standardreceived both a 10 percent railroadre-
bate plus a commission paid by the railroadsto their gatheringpipeline
company, the AmericanTransferCompany.The railroadspaid this com-
mission on all barrels shipped by the pipeline, not just on Standard's
barrels." The total effect was to give Standarda substantialtransporta-
tion cost advantageover its refiningrivals. For example, while the open
rate in 1878 for crude shipments to New York was $1.70 per barrel,
Standardpaid only $1.06 per barrel.52Standard's $1.06 transportation
rate implied a marginalrefineryprofit of plus $.49 per barrel, while the
independents'$1.70 transportationrate implieda marginalrefineryprofit
of minus $.15 per barrel, or 9 percent of the gross rate.53The indepen-

49Blanchardtestimony, HepburnHearings,supra note 3, at 3492.

50 Coal and grain were categorizedas class 4 commodities,while oil was classified as
class 3 until 1886.These classificationsare arbitraryin the sense thatthey do not correspond
to differentper-ton-miletransportationcosts. The dates for this and later analysis were
chosen solely on the basis of the availabilityof consistent data. Unfortunately,we have
not been able to find a coal rate for 1872or 1874.
5' These commissions varied from 20 to 35 cents per barrel. See Cassatt testimony,
Trusts,supranote 1, at 186-89. In a February1878letter betweenO'Day, generalmanager
of the AmericanTransferCompany,and Cassatt, vice presidentof the PennsylvaniaRail-
road, O'Dayjustifiespaymentof these commissionsby writing,"The fruitof co-operation
referredto has been fully evidencedin the fact thatsince last fall yourcompanyhas received
fifty to sixty cents perbarrelmorefreightthanwas obtainedby it priorto our co-operation."
See letters between Cassatt and O'Day, dated February15, 1878, and May 15, 1878, in
1 Tarbell,supra note 6, at 374-75.
52 Williamson& Daum,
supra note 1, at 428.
53 Marginalrefineryprofitis definedto equalthe refinedproductprice minuscrude price
minus marginalrefinerycosts and transportationcosts. In 1878the refinedproductprice
was $4.52 per barrel,the crude price (dividedby .65 to adjustfor refinedyield) was $1.71
per barrel, and refiningcosts are estimatedto be $1.26 per barrel,for a refiningmargin
before transportationcosts of $1.55 per barrel.(The refinedprice and the crude price are
the averageyearly prices as reportedin Derrick's,supra note 41, at 783, 711, adjustedto
42 gallon barrels).The marginalrefiningcosts (excludingany capitalcosts) are estimated
by extrapolatingbetween the 1870cost of 5 cents per gallonand the 1880cost of 21/2cents

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 21

dents' refining losses caused by these differentialtransportationrates

enabledStandardto acquirethe remainingrefinersin Pittsburgh,Philadel-
phia, New York, and the Oil Regions.
Similarto Rockefeller's use of the South ImprovementCompanycon-
tract in Cleveland, the evidence indicates that many of the independent
refiners acquired during this period were squeezed by the unfavorable
transportationrates they faced and, as a consequence, sold their refiner-
ies at distress prices. Cases of such distress sales include, for example,
the acquisitions of A. H. Tack (a refinerin Pittsburgh)in 1874 and Mr.
Harkness (a refinerin Philadelphia)in 1876. Both refinerstestified at the
House Committeeon ManufacturesTrust Investigationin 1888that they
sold out because of a squeeze on the part of railroads.54Rockefeller
attemptedto keep Standard'srole in these and other acquisitionsin these
cities hidden by using as "fronts" his originalpartnersin the South Im-
provement Company that he had secretly acquired in 1874. From the
independentrefiners'perspective, it was the railroadsthat increasedtheir
shippingrates and squeezed their margins,while the local refinerswhich
bought them out helped them salvage something from their business.
The independentrefinerswere unawarethat they were selling out to the
Rockefellerintereststhat were responsiblefor the highershippingrates.5
In 1873there were 22 independentrefineriesin Pittsburgh;by the end
of 1877 Standardand its affiliateshad acquiredall of them.56Similarly,
refineryacquisitionsin the Oil Regionsbeganin October 1875,when John

per gallon, multipliedby 42 to convert to barrels(Williamson& Daum, supra note 1, at

230, 483).
54Testimonyof AugustusH. Tack and of WilliamW. Harknessin Trusts, supra note 1,
at 214-15 & 223-25. See also 1 Tarbell,supra note 6, at 155, 157.
55 Rockefellermay have kept his role secret because of the fear that knowledgeof his
connection with the railroadswould lead to political pressure or to legal challenges (as
had occurredwith the South ImprovementCompany,supra note 36). Legal challengesto
Standard'sactions did occur in August 1878when the PetroleumProducers'Union con-
vinced the Commonwealthof Pennsylvaniato bringsuits againstthree railroadsthat oper-
ated withinPennsylvania(the Pennsylvania,AtlanticGreatWestern,and Lake Shore)and
againstStandardOil's UnitedPipelinefor failingto properlyexecute theirduties as common
carriers.In addition, in August 1879the PetroleumProducers'Union convinced a grand
jury, using evidence from the commoncarriercase, to indict seven StandardOil officials
for conspiringto restraintrade. In February1880Standardsettled these suits by makinga
contract with the PetroleumProducers'Union. This was followed by an agreementthe
PetroleumProducers'Union made with the PennsylvaniaRailroadin April 1880 which
stipulatedthat there would be no rate discrimination.It was agreed that large shippers
would be given a "reasonable"and public advantagebut that any person shippinga like
quantitywould pay the same rate. However, by this point in time only Standardwas large
enough to obtain the most favorablerates (Williamson& Daum, supra note 1, at 431-33;
1 Tarbell,supra note 6, at 241-62, 401-6).
56 Henry, supra note 2, at 318; 1 Tarbell,supra note 6, at 161.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 TABLE 3
RAILRATESFORVARIOUS 1872-84 (Cents per Ton Mile)
COMMODITIES,

FROMOIL REGIONS FROMLEHIGH

TO NEW YORK VALLEY TO RAIL RATES FORP
($) NEW YORK: FROM CHICAGO RELATIVETO G
ANTHRACITE TO NEW YORK:
TIME Refined Crude COAL GRAIN Refined Crude
PERIOD Products Oil ($) ($) Products/Grain Oil/Grain
1872* .0168 .0151 . .0122 1.38 1.24
1874t .0213 .0185 .? . . .0109 1.94 1.68
18775 .0270 .0163 .0089 .0097 2.77 1.68

1880? .0089 .0085 .0104 .0073 1.23 1.17

188411 .0074 .0064 .0105 .0073 1.01 .88
SOURCE.-Rates in terms of cents per hundred pounds are converted to cents per ton mile by dividing by 477 miles for
City to New York), by 822 miles for grain rates (the distance from Chicago to New York), and by 150 miles for coal rat
to New York) and then by multiplying the rates by 20. In converting per-barrel oil rates to tons, a barrel is assumed to eq
is in InterstateCommerceCommission,Railwaysin the UnitedStates, PartII, at 87 (1903)(hereafterICC).
* Oil rates for September25, 1872,are from4 RailroadGazette 144(1872);grainrates for September1, 1872,are from
t Oil ratesfor October24, 1874,are from6 RailroadGazette417 (1874);grainrates for May 16, 1874,are from 6 Rail
t Oil rates for December1, 1877,are from ICC, at 87; grainrates for December28, 1877,are from 9 RailroadGazet
1877,are fromICC, at 88.
? Oil ratesfor April 1, 1880,are from ICC, at 87; grainratesfor May 14, 1880,are fromICC, at 79; coal rates for Ap
1 Oil ratesfor February1, 1884,are from ICC, at 87; grainrates for January14, 1884,are fromICC, at 79; coal rates

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 23

D. Archboldformedthe Acme Oil Companyand agreedto join Standard,

bringingthe first-and third-largestrefineriesin the Oil Regions underthe
control of Standard."Following this amicablemergerwith Standard,the
Acme Oil Company quickly purchaseda large numberof other Oil Re-
gions refineries. In 1875 Standardalso acquireda large numberof New
York refiners.From 1876through1879Standardcapitalizedon its highly
advantageoustransportationrates to continue its acquisition of the re-
mainingindependentrefineries,gainingcontrol of an additional108firms
or properties. A summaryof Standard'srecords shows that "24 were
bought or leased in 1876, 35 in 1877, 40 in 1878and 9 in 1879."58

IV. MONOPOLIZATION BY "RAISING RIVALS' COSTS"

The evidence is overwhelmingthat Rockefellerexpanded his position

in refiningby "raisingrivals' costs," in the sense that Rockefeller used
his substantialtransportationcost advantageto acquiremany of his refin-
ing rivals at distress prices.59In addition,we describe below how Rocke-
feller used his transportationcost advantageto prevent the entry of new
refiners. However, while preferentialrail rates did disadvantageRocke-
feller's actual and potential refiningrivals, merely labeling these events
as an example of "raisingrivals' costs" does not explain what occurred.
To understandwhat occurredone must answer three questions: (a) How
was the petroleum industry monopolized?(b) How was the monopoly-
monopsony profitcreated by the monopolizationof the petroleumindus-
try shared between Standardand the railroads?and (c) Which stage of
the petroleumindustry,transportationor refining,actually was monopo-
lized?
The answer to the first question is that the petroleum industry was
monopolized by joint action of Standardand the railroads. Given that
the three stages of petroleum industry production-crude production,
refining, and transportation-are complementary, a monopoly at any
stage would monopolize the entire industry. However, as we have seen,
no firm or group of firms acting on any single stage could monopolize
that stage. Neither the crude producersnor the refinerscould create a
monopoly because both of these stages had an extremely large number
of firms and easy entry. Transportationhad only three firms and signifi-
cant costs of new entry, but each railroad'slow marginalcost and large

57 1 Nevins, supra note 16, at 212.

58 Fromid. at 422 n.2, citingTaylor'sunpublishedHistoryof the StandardOil Company.
59 The term "raisingrivals' costs" was first used, in a differentcontext, in Steven C.
Salop & David T. Scheffman,RaisingRivals' Costs, 73 Am. Econ. Rev. 267 (1983).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
24 THE JOURNAL OF LAW AND ECONOMICS

excess capacitymadeit impossiblefor themto collude successfully. What

our analysis indicates is that Standardand the railroads,by cooperating
with one another, did somethingjointly that neither of them could do
separately-they created a monopoly. In particular,they jointly estab-
lished a transportationcartel, and the railroadsfacilitated Standard's
growth so that Standardcould effectively police the cartel with its domi-
nant position in refining.
The answer to the second question is that Standardand the railroads
shared the monopoly profits created by theirjoint effort based on their
relative bargainingpower. From Figure 2 the total per-barrelindustry
monopsony-monopoly profit is Pm - E. If the railroads had sufficient
bargainingpower so that they did not have to share any industryprofits
with Standard,they would receive rail rates of Pm - D and, net of their
marginalcosts of E - D, earn for themselves the entire per-barrelprofit
of Pm - E. Alternatively, if Standard had sufficient bargaining power so
that they did not have to share any industry profits with the railroads,
rates received by the railroadswould equal E - D, the marginalcost of
rail transportation, and Standard's profit would equal Pm - E, the price
chargedby Standardfor refinedkerosene minus the sum of their cost of
crude, their cost of refining,and their cost of transportation.Bargaining
by the two parties set the rail rates, and, therefore, the actual profit
shares received by the railroadsand by Standardsomewhere between
these two rail rate benchmarks of PM - D and E - D.60
A measure of Standard'sshare of industryprofitcan be calculatedby
comparingthe rebates received by Standardon their own shipments(and
the drawbacksthey received on the rates paid by competitors)with the
railroads'increased receipts. For example, using the refined rates from
Clevelandto New York presentedin Table2, Standard'srebateunderthe
South ImprovementCompanycontractwas $.50 per barrel.The railroad
receipts increasedfrom the $1.30 per barrelnegotiatedrate in 1870to the
$1.50 net rate under the South ImprovementCompanycontract, or $.20
per barrel. Therefore, Standardappearsto be earning71 percent ($.50/
$.70) of the total increase in profit. However, this is clearly an overesti-
mate of Standard'sprofit share.
First of all, this calculationoverestimates Standard'sprofit share be-
cause only a portionof the rebate is passed on in a higherrefiningprice.

6" We can assume that even when the railroad rate paid by Standard is substantially
higher than the lower benchmark of E - D, it does not affect how Standard priced the
refined product, as summarized in Figure 2. In order to maximize total industry profit,
Standard should always demand crude and sell kerosene as if the rail rate were E - D,
with any increase in the actual rail rate above this level merely representing how the
railroads collected their share of collusive profits.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 25

As we have seen, in orderfor Standardto acquirethe dominantshare of

refining that was necessary for cartel policing, transportationrates to
independentrefinershad to be set above the monopoly rate of Pm - D.61
This created the negative independentrefinerprofitsthat were necessary
to drive independentrefinersout of the marketand to discouragepoten-
tial entrants. Therefore, Standard'srebate from the gross transportation
rate overstates their profit. Using 9 percent of the gross rate as an esti-
mate of the negative marginalprofits earned by independent refiners62
and, hence, as an estimate of the overstatementof the level of Standard's
profit in the rebate, would lower Standard'sprofitby $.18 (.09 x $2.00)
and Standard's profit share to 62 percent ($.32/$.52). In addition, the
$1.30 benchmarkrate is an overestimateof the average competitive rate
that existed precartel because it does not take account of the periodic
rate wars that occurred.Accountingfor this would raise the total industry
profit attributedto the cartel, thereby lowering Standard'sprofit share
further.
This analysis highlightsthe distinction between the profit-sharingas-
pects and the policing aspects of this cartel. Disadvantageousrail rates
paid by the independent refiners should be thought of as a necessary
condition to preserve the railroadcartel policing asset, not as the way in
which petroleumindustryprofitswere sharedbetween Standardand the
railroads. If Standard could have policed the railroad cartel without
achieving dominance in refining,disadvantageousrates would not have
even been necessary. The railroadscould have charged all refiners, in-
cludingStandard,the same monopolyrate, Pm - D, and made a separate
side payment to Standardfor its policing effort. It is in this sense that
the type of "raisingrivals' costs" that occurredin the StandardOil case
is differentfrom the usual type of "raisingrivals' costs" discussed in the
recent economic literature.
In the usual "raisingrivals' costs" case, a vertical relationshipis said
to create an incentive to increase the price (or "foreclose" the supply)
of an input to rivals which, in turn, creates a cost disadvantagefor the
rivals sufficientto allow the firmto exercise increasedmarketpower. In
particular, a vertical connection, without any horizontal conspiracy,
leads an upstreamfirmto restrictthe supply of its outputto downstream

61 The magnitudeby which independentrates could exceed Standard'srates was gov-

erned by legal constraints.Under state common carrierstatutes, independentrefinerrail
rates could not legally be set unreasonablyhigher than Standard'srates, nor could the
railroadshave exclusive dealingarrangementswith Standard.See note 55 supra.
62This estimateof the loss does not includeany returnon capitalinvestedby independent
refiners.Sew note 53 supra.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
26 THE JOURNAL OF LAW AND ECONOMICS

rivals, thereby raisingthe price of the final output.63In the StandardOil

type of case, in contrast, the vertical relationshipdoes not change the
incentive to restrict supply; it merely facilitates collusion among the up-
stream firms.64
The answer to the thirdquestionregardingwhat stage of the petroleum
industrywas monopolizedis not refining,the answer implicitin the usual
descriptionof the StandardOil case. Given the fact that actions on both
the refiningand transportationlevels were requiredto create the monop-
oly, it certainly does not make sense to isolate refiningto the exclusion
of transportation.It also is misleadingto describe the situationas if mar-
ket power was created in refiningwith the assistance of the railroads.
Refiningwas not monopolizedsince a new entrantinto refiningcould not
expect to earn any monopoly profit; a new entrant into transportation,
however, could expect to earn monopoly profit and to destabilize the
cartel. Moreover, given the ease of entry into refiningand Standard's
behaviorin stabilizingeach railroad'sshareof shipments,it is more accu-
rate and insightfulto describe the situationas one where marketpower
was created on the transportationlevel with the assistance of Standard.
What Standarddid was replicatethe situationthat would have existed
under the South ImprovementCompanyarrangement.That is, Standard
monopolizedpetroleumtransportationand sharedthe resultingmonopoly
profit with the railroads. The only analytical difference between what
Standardaccomplishedand what would have existed underthe proposed
collusive South ImprovementCompanycontractis that, instead of being
one of the evening refiners"hired" by the railroadsin 1871to police the
collusive transportationagreement, Standardnow essentially "owned"
the policing asset. Standardcollected monopoly profits for its cartel-

63 See, for example,JanuszA. Ordover,GarthSaloner,& Steven C. Salop, Equilibrium

VerticalForeclosure,80 Amer. Econ. Rev. 127(1990),for a formalstatementof this effect.
64 This type of case is colorfullylabeledby Krattenmakerand Salop in their elucidating
taxonomyof "raisingrivals'costs" cases as "TheCartelRingmaster"(ThomasG. Kratten-
maker & Steven C. Salop, AnticompetitiveExclusion:RaisingRivals' Costs to Achieve
Power over Price, 96 Yale L. J. 209 (1986)).Krattenmaker& Salop (at 240) recognizethat
the CartelRingmastertype of "raisingrivals'costs" case "has a greaterhorizontalaspect"
than the other types of "raisingrivals' costs" cases they discuss. However, they claim
elsewhere that horizontalmarketanalysis alone is insufficientto understandwhat is oc-
curringin a CartelRingmastertype of case (ThomasG. Krattenmaker& Steven C. Salop,
Competitionand Cooperationin the Marketfor ExclusionaryRights, 76 Am. Econ. Rev.
Papers& Proc. 109, 112(1986)).They argue,in part,that profitsearnedby the downstream
firm can be used to make side paymentsto the upstreamcolludingfirmsto stabilize the
cartel. However, while some profitsearnedby Standardcan be thoughtof as being trans-
cartel,the movementfromcompe-
ferredbackto the railroadsto stabilizethe transportation
tition to monopolyalways leads to the creationof profitsthat can be used as bribesto pay
transactorsoff. The existence of such monopolyprofitshas nothingto do with the vertical
or "raisingrivals' costs" aspect of the cartelenforcementproblem.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 27

policing services directly from the crude suppliers and the consuming
public, paying a monopsony price for crude and charginga monopoly
price for refinedkerosene; but, as we shall see in the following section,
Standard's profit vanished whenever the transportationcartel broke
down.
The collusive transportationscheme createdby Standard,in principle,
could have been established without Standardeven being a refiner. For
example, Standardcould have policed the railroads' collusive pooling
arrangementby operatinga clearinghousethroughwhich all petroleum
shipmentscould have moved and, in return,received a share of the mo-
nopoly transportationprofits. This alternative collusive arrangement
would have been less effective thandirectpolicingby a dominantevening
refiner. But one must not confuse what Standardaccomplished, namely,
creation of a transportationcartel, with how Standardaccomplished it,
by using its dominantposition in refiningas an effective policing device.
Further,what Standardaccomplishedand how Standardaccomplishedit
are distinct from how much Standardreceived as compensationfor its
efforts. Although Standardearned a significantshare of industryprofits
on its dominant refining operations, our analysis indicates that it was
petroleumtransportation,not refining,that was monopolizedand that the
profits earned by Standardin refiningshould be thought of as merely a
share of the monopoly profits from the transportationcartel. Only by
focusing on what was monopolized,not who was doing the monopolizing
or who was collecting most of the monopoly return, can we understand
the natureof StandardOil's marketpower.

V. STABILITY OF THE CARTEL

Stabilityof the transportationcartel establishedand enforced by Stan-

dard hinged on Standard'sability to punish effectively any railroadthat
cheated on the collusive agreement by granting independent refiners
transportationrebates and on Standard'sability to prevent the entry of
transportationalternatives to the three colluding railroads. Although
Standardpaid the railroadswell, certainlyby comparisonwith the rates
that existed duringthe period of "ruinous"competitionand by compari-
son with the rates that were chargedat the time on other products, sig-
nificant tension existed among the railroadson how shipments would
be shared, and between the railroadsand Standardon the allocation of
monopoly profits. This led at times to a railroad(most frequently the
Pennsylvania)loweringits rates to independentrefinersand breakingthe
collusive agreement.As we shall see, Standardalways respondedaggres-
sively to these violations of the collusive agreementand, in addition,also

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
28 THEJOURNALOF LAWAND ECONOMICS

responded aggressively to potential entrants into petroleum transporta-

tion that could destabilize the cartel.

A. The Empire Rate War

The most dramaticexample of a breakdownin the collusive arrange-
ment occurredduring1876-77. In 1876the EmpireTransportationCom-
pany, the firm associated with the Pennsylvaniathat coordinatedtraffic
on the rail lines moving oil from the Oil Regions to New York, entered
into refining.65By verticallyintegratinginto refiningon a significantscale,
the Pennsylvaniacould now evade the railroadcartel by lowering effec-
tive rates to the Empirerefineries,therebyshiftingcrude shipmentsaway
from the other railroadsto the Pennsylvaniaand away from Standard
refineriesto the refineriesserved by the Empire. Once the Empire's re-
finery operationsgrew, Standardwould not have been able to effectively
punish the Pennsylvania.66Therefore,Rockefellerhad to respond to this
challenge if the railroadcartel was to survive.
In early 1877Rockefellerentered into discussions with the Pennsylva-
nia Railroadto prevent furtherexpansion of the EmpireTransportation
Company into refining. When these discussions brought no results, in
March 1877 Standardwithdrewall its shipmentsfrom the Pennsylvania,
shutting down its Pittsburghrefineries and runningall its crude at its
Cleveland refineries.67Oil exports from Philadelphiaand Baltimore, the
Pennsylvania's main ports, decreased dramatically,and exports from
New York correspondinglyincreased.68In response, the Pennsylvania
dramaticallycut its rates to crude producers and independentrefiners
and initiateda price war among all three railroads.69
65In October 1876the EmpireTransportationCompanypurchasedtwo refineries,one
in New York and one in Philadelphia.In early 1877the Empiretook steps to enlargeits
New York refineryand began constructionof a new large refineryin Philadelphia.See
Cassatt testimony, Trusts, supra note 1, at 175, 178.
6 The Empire'srefinerieshad a capacity of 4,000 barrelsper day (Nevins, supra note
16, at 240). This amountedto approximatelyone-thirdof the Pennsylvania'sallocatedshare
of petroleum shipments. Moreover, the Empire was engaged in furtherexpansion, the
result of which would have severely limited the sanction Standardcould impose on the
Pennsylvania.
67 Cassatttestimony, Trusts, supra note 1, at 175; 1 Nevins, supra note 16, at 239.

8 The share of kerosene exportsfrom the ports of Baltimoreand Philadelphiafell from

45 percentto 25 percent(in January-Mayof 1877comparedto January-Mayof 1876).Over
this periodthe exports in New York increasedfrom 52 percentto 71 percent.The share of
exports from Philadelphiaalone, which had fallen to 15 percent duringJanuary-Mayof
1877, fell to only 9 percent duringJuly and August (RailroadGazette, June 15, 1877, at
271, and September21, 1877,at 431).
69 In the springand summerof 1877,duringthe EmpireTransportation ratewar, the Erie
Railroadlowered its crude rate to the seaboardto 35 cents per barrel.This rate was less
than one-thirdof theirrate in March1877(BlanchardTestimony,HepburnHearings,supra
note 3, at 3462;Williamson& Daum, supra note 1, at 424.)

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 29

The Empire Rate War is clearly reflected in market prices. Figure 3,

which plots crude and refined prices in the United States from 1869 to
1905,indicatesthat both refinedprices and crudeprices declined dramati-
cally duringthe decade of the 1870s when Rockefeller was establishing
Standard'sdominantposition in refining.Refined prices dropped by 75
percent, from 33 cents per gallon in 1869 to 8 cents per gallon in 1879,
and crude prices droppedby 85 percent, from 13 cents per gallon in 1869
to 2 cents per gallon in 1879.The only aberrationin this downwardtrend
is the upwardspike in crude and refinedprices that occurs in 1876-77, a
period that includes the EmpireRate War.70
While crude producersand refinersprofitedsubstantiallyby this rate
war, Standardhad much to lose. If the rate war continued, Standard's
profitfrom its position as a railroadcartelpolicer would vanish. On Octo-
ber 17, 1877, the Pennsylvania,which suffered drastic losses along with
the Empire duringthe rate war, was finally driven to exercise its option
to buy the Empire and to make peace with Standard.71In late 1877 the
railroadsand Standardestablished a new rate agreement,with Standard
once again enforcing these rates by evening traffic. The agreement set
each railroad'sshare as follows: the PennsylvaniaRailroad,47 percent;
the Erie Railroad,21 percent;the New York Central,21 percent; and the
B&O, 11 percent.72In exchange for enforcing these market shares and
thereby assuring the railroadsthat their rates would not fall to ruinous
levels, Standardcontinuedto receive a 10 percent preferentialrebate on

70 The doublingof crude prices during 1876-77 was not entirely due to the breakdown
of the monopsonisticrailroadcartelduringthe EmpireRate War. In addition,there was an
unplanneddeclinein crudeproductionrelativeto demand,reflectedin a significantdecrease
in the average yield per well and a decline in inventory levels. (See David D. Leven,
PetroleumEncyclopedia:"Done in Oil" 40 (1941).Also see S. C. T. Dodd, Combinations:
Their Uses and Abuses with a History of the StandardOil Trust 3 (1888).) However,
increasedrail competitionclearly reinforcedthese supply-sidefactors. This can be seen in
1877,when crude prices remainedhigh as refinedprices were falling.
71 Standardfinancedthe settlement, purchasingthe EmpireTransportationCompany's
oil-relatedassets, includingits tank cars, refineries,and pipeline network, and makinga
loan to the PennsylvaniaRailroadso it could purchasethe Empire'sgeneralmerchandise
cars (RailroadGazette, November9, 1877, at 499; Cassatt testimony, Trusts, supra note
1, at 180).
72 Standarddetected deviationsfrom these shares by receivingextensive marketintelli-
gence fromits agents, who closely monitoredoil shipmentsthroughoutthe system by direct
observation and by arrangementswith freight office clerks and competitors'employees
(2 Tarbell, supra note 6, at 35-38). Accordingto one competitorat the time, Standard
"keep[s] a very accuraterecordof every barrelthat the independentrefinersput into each
city, town, and hamlet" (testimonyof Theodore B. Westgate, 1 IndustrialCommission,
PreliminaryReporton Trustsand IndustrialCombinations697-98 (1900)(hereafterIndus-
trial Commission)). In addition, beginningin November 1877 the railroadsexchanged
monthlystatementson the volumeof oil shipmentsandbroughtany deviationsto Standard's
attentionfor correctionthe followingmonth. See Cassatttestimony,Trusts, supra note 1,
at 184;Nevins, supra note 16, at 248-49.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 35

30

25

20

6 15

10

O
00
rI
00
r00 r00 r
00
Ir-
00
00 00 00 00 00 O O O
00 00 00 00 00 00 00 00

FIGURE 3.-Oil prices in the United States, 1869-1905. Source: Annual data from U.S. Bureau o
Commissionerof Corporationson the PetroleumIndustry,Pt. 2, at 622-23 (2 Pts. 1907),table 165. (P
on a gold basis.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 31

its own shipmentsand the guaranteethat the rates it paid would be lower
than its rivals.73It was duringthis period, immediatelyafter the Empire
agreement in 1877, that Standardconsolidated its hold on the industry
and the primarythreat to the transportationcartel became the entry of
new transportationfirms.

B. Entrants into Petroleum Transportation

The incentive for new firmsto enter petroleumtransportationwas de-
terminednot by the profitbeing earned by the railroads,but by the total
industryprofitbeing earnedby Standardand the railroadscombined(rep-
resented by the per barrelprofitof P, - E in Figure2). While new entry
into refiningremained effectively deterred by the disadvantageousrail
rates set by the cartel, any new transportationfirm knew that if they
supplied a new independent transportationlink, independent refiners
would enter to use the link and that they would not have to share their
profitwith Standard.
The only feasible potential railroadentrantinto petroleumtransporta-
tion was the B&O Railroad, which ran from Baltimore to Columbus,
Ohio, via Wheeling,West Virginia.74 The B&Oinitiallyattemptedto enter
the petroleumtrade in 1871when it built a branchline from Wheelingto
Pittsburgh.75This smalladditionto its railroadsystem enabledit to supply
transportationservices to the coast for Pittsburghrefiners. However,
despite expectations that a large oil traffic would be diverted from the
Pennsylvania's Pittsburgh-Philadelphia route to the B&O's Pittsburgh-
Baltimoreroute, in 1874Baltimoreaccountedfor only 2.3 percent of U.S.
exports.76 The Pennsylvaniacontrolledrail transportationof crude from
the Oil Regions to Pittsburghand apparentlyused this to requirerefiners
in Pittsburghto move their refinedproductto the seaboardon the Penn-
sylvania.
The B&O became an effective competitor of the Pennsylvania only

73 Cassatt testimony, Trusts, supra note 1, at 190. In late 1877and early 1878all three
railroadsalso began payingcommissionsto Standard'sgatheringline company,the Ameri-
can TransferCompany.(See discussionat note 51 supra.)
74 Because the petroleumtrade was only about 4 percentof total railroadtrafficin tons
on, for example, the PennsylvaniaRailroadin 1871-72 (risingto 6 percentin 1873-74 and
7 percent in 1874-75), it was not economicto makelargerailroadsystem investmentsjust
to serve the profitablepetroleumtrade. Only railroadsthat had an existing trunkline near
the Oil Regions, such as the three existing cartel membersoriginallyhad, would find it
profitableto enter. See Henry V. Poor, Manualof the Railroads409 (1871-72), 563 (1873-
74), 450 (1874-75).
75 Railroadsin the NineteenthCentury22 (RobertL. Frey ed. 1988).
76 RailroadGazette, September15, 1876,at 407.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
32 THEJOURNALOF LAWAND ECONOMICS

when the ColumbiaConduitCompanycompleted a pipelinefrom the Oil

Regions to Pittsburghin 1875 and began delivering crude oil to Pitts-
burgh.77The Pennsylvaniavigorouslyfought constructionof the Colum-
bia Conduit, claimingpipeline constructionwould damage its track and
refusingto permitthe pipeline to cross its right-of-way,notwithstanding
the Conduit's offer to post a damage bond.78 The Conduit began deliv-
ering crude oil to Pittsburghonly by unloadingthe oil and carryingit in
wagons on the public roadway across the Pennsylvania's right-of-way.
In spite of these disadvantagesthe B&O-Conduitcombinationoffered
Baltimorerefinersa total throughrateto move crudefromthe Oil Regions
to their refineriesof only $1.15. Less than 6 months later in late 1875,
after the Pennsylvaniaand Standardreached agreementwith the B&O
and the B&O was broughtinto the Standardenforced railroadcartel, the
open tariffwas increasedto $1.88.79 To police B&O shipments,Standard
acquiredrefineriesin the areas served by the B&O and in the fall of 1877
purchasedthe ColumbiaConduitCompany.80
Entry of another transportationalternativeto the now four colluding
railroadsoccurred in 1879 when the TidewaterPipeline Company con-
structed the first pipeline to transportcrude from the Oil Regions over
the mountainsto Williamsport.81 (See map, Figure 1 above.) From Wil-
liamsport the oil then was transportedeitherto Philadelphiaor New York
by the Reading Railroad and its connections.82 In this way the Tidewater-
Reading combination became a transportationalternativefor independent
refiners, permittingthem to circumvent Standard'scontrol of transpor-
tation.83
Standardtried to block constructionof the Tidewaterpipeline by ac-

77 In 1875Baltimore'sshare of exports increasedto 11.7 percent. See RailroadGazette,

September15, 1876,at 407.
78ArthurMenziesJohnson,The Developmentof AmericanPetroleumPipelines:A Study
in PrivateEnterpriseand PublicPolicy, 1862-1906,at 35, 40-41 (1956).
79Williamson& Daum, supra note 1, at 409.
80See Williamson& Daum, supra note 1, at 419; Tarbell,supra note 6, at 195.
81 Althoughthe ColumbiaConduitwas the firstlong-distancecrudepipeline,it essentially
flowed downhillfrom the Oil Regions to Pittsburghand, therefore,did not have to solve
the pumpingproblemsinvolved in transportingoil over the Allegheny mountainsto the
seaboard. The Tidewaterwas the first long-distancepipeline to solve this technological
problem.See Johnson,supra note 78, at 75.
82MarvinW. Schlegel,Rulerof the Reading:The Life of FranklinB. Gowen, 1836-1889,
at 182 (1947);Williamson& Daum, supra note 1, at 444.
83 Because it was believed until 1893that refinedproductswould be damagedif shipped
by pipeline,the Tidewater'sshipmentsconsisted solely of crudeshipments.The Tidewater
delivered crude oil to existing independentrefinersin Philadelphia,to one independent
refinerylocated at Williamsportand to two new refineriescompleted by the Tidewater
PipelineCompanyitself. See Williamson& Daum, supra note 1, at 443, 450.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 33

quiring a "deadline," or blocking right-of-way,across the Tidewater's

planned pipeline route from the Oil Regions to Williamsport.The rail-
roads cooperated with Standardin this effort by putting their rights-of-
way at Standard'sdisposal.84One would expect the railroads to work
against the Tidewater's entry because a long-distancepipeline is a low-
cost substitutefor what the railroadssupply. However, because a long-
distance pipeline is a low-cost substitutefor an importantinput that re-
finers demand, one would not expect refiners to work against the
Tidewater's entry. Refiners, especially a dominantrefinersuch as Stan-
dard, would ordinarilyperceive the entry of the Tidewater as a very
favorable development. Therefore, Rockefeller's behavior in attempting
to block the Tidewater is furtherevidence that he was not acting as a
monopoly refiner,but as a policingagent of a railroadcartel. Rockefeller
correctly perceived that once the railroadslost their collective market
power, Standardwould lose its share of the monopoly rents from the
petroleumtransportationcartel.
The attemptby Standardand the railroadsto block constructionof the
Tidewater pipeline failed. Despite the fact that two railroadshad to be
crossed, the Tidewaterfound a gap in the deadlinein partby runningthe
pipeline under a railroadin a dry creek bed, giving the pipeline a clear
route to Williamsport.85 The successful entry of the Tidewatercaused a
rate war between the pooled railroadsand the Tidewater-Readingcombi-
nation beginningin June 1879.86As could be expected, by early 1880the
TidewaterPipeline was invited to join the railroadcartel, and the parties
tentatively agreedto raise rates. The proposedcartel agreementincluded
a new rate structureand marketshares were set for all the transportation
alternatives, with the Tidewatergiven 11.5 percent of oil shipments, ap-
proximatelyequal to the pipeline's capacity.87
Table 4, which summarizes the available evidence on collusively
agreed-on market shares for the petroleum transportationalternatives,
indicates that the 1880agreementwith the Tidewaterlowered the Penn-

84 2 Tarbell,supra note 6, at 4; Williamson& Daum, supra note 1, at 441.

85 Johnson,supranote 78, at 75. The Tidewateractuallyappliedfor and obtaineda patent
on its route. The legal precautionstaken by the Tidewaterto protectits right-of-waysaved
it fromthe violence and costly litigationthat plaguedconstructionof the ColumbiaConduit
pipeline. See Williamson& Daum, supra note 1, at 442.
86 The rate war caused the rate to the seaboardfrom the Bradfordfield, a newly discov-
ered field to the northof the older Oil Regionsfields, to fall from 85 cents per barrelto 15
cents per barrel(Schlegel, supra note 82, at 183).
87 The new open ratefromBradfordto the seaboardwas set at 60 cents per barrel,below
the previousopen railrateof 85 cents but well above the ratewarlevel of 15cents (William-
son & Daum, supra note 1, at 445-46).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
34 THE JOURNAL OF LAW AND ECONOMICS

TABLE 4
COLLUSIVEPETROLEUMTRANSPORTATION
SHARES(%), 1871-84

TRANSPORTATION
ALTERNATIVE

B&O/ Tidewater/
DATEOFAGREEMENT Pennsylvania Erie Central Conduit Reading
1871(South Improvement
Companyagreement) 45 27.5 27.5
1874(RutterCircularagree-
ment) 50 25 25
1875(B&Oagreement) 52 19.5 19.5 9
1877(Empireagreement) 47 21 21 11
1880(proposedTidewater
agreement) 27 N.A. N.A. N.A. 11.5
1883-84 (finalPennsylvania/
Tidewateragreement) 26 N.A. N.A. N.A. 11.5
SOURCE.-For 1871 see W. G. Warden, Petroleum Producers' Union, A History of the Rise and Fall
of the South Improvement Company 98, 103, 109 (1872). For 1874 see 1 Allan Nevins, Study in Power:
John D. Rockefeller, Industrialist and Philanthropist 197 (2 Vols. 1953). For 1875 see the testimony of
A. J. Cassatt in the 1879 case Commonwealth v. The Pennsylvania Railroad Company; reported in U.S.
House of Representatives, Committee on Manufactures, Report on Investigation of Trusts (H.R. 50th
Congress, Ist Sess., H.R. Rep. No. 3112, at 199 (Washington, 1888) (hereafter Trusts)). For 1877 see
Cassatt testimony, reported in Trusts, supra, at 184; 1 Nevins, supra at 248-49. For 1880 see Harold
F. Williamson & Arnold R. Daum, The American Petroleum Industry: The Age of Illumination, 1859-
1899, at 451 (1959). For 1883, see the agreement between Standard and Tidewater Pipeline reprinted in
2 Ida M. Tarbell,The History of the StandardOil Company303 (1904). For 1884see the agreement
between Standard and the Pennsylvania Railroad reprinted in 2 Tarbell, supra, at 309-10.
NOTE.-N.A. = data not available.

sylvania's share of petroleum transportationby 20 percentage points

compared to the previous Empire agreement.The Pennsylvania, there-
fore, began cheating on the proposed Tidewateragreement, attempting
to increase its traffic by offering a new lower open rate to all refiners
and inducing several new independentrefinersto enter.88In response,
Standardcut its shipmentson the Pennsylvania,and the Pennsylvania's
share of oil shipmentsdroppedfrom the 27 percent stipulatedunder the
proposed cartel agreementto only about 13 percent.89A stable collusive
agreementon marketshares and rail rates, which remainedin effect and
unchanged for 18 years, was finally reached in 1884, after Standard

88 Testimonyof Lewis Emery,Trusts,supranote 1, at 235-36. The Pennsylvaniareduced

its rate from the Bradfordfield to 33 cents from the agreed-on60 cents. See Williamson&
Daum, supra note 1, at 446, 451. (We would not expect the Pennsylvaniato attempt to
commitcrediblywithregardto these ratesto independentrefinersbecausethe Pennsylvania
does not want the cartel to break down; presumably,such long-termcommitmentswere
unnecessary to induce independentrefinerentry because of the relatively low level of
nonsalvageablerefinerinvestments.)
89 Williamson& Daum, supra note 1, at 451.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 35
shifted the bargainingpower in its favor by completing construction of
its own long-distancepipeline network.

C. Standard's Constructionof a Pipeline Network

Standardbegan constructionof its own long-distancepipeline network
immediatelyafter the Tidewater'ssuccess. Until the Tidewater'stechno-
logical advance in pipeline construction,new entry into petroleumtrans-
portationremainedeffectively blocked. The four rail lines that could en-
ter economicallyby buildinga spurinto the Oil Regions had alreadydone
so and were part of the conspiracy;it did not pay anyone else to build a
new railroad solely to service the oil trade. Once long-distance crude
pipelines became technologicallyfeasible, however, the railroadcartel
was threatenedwith the possibilityof significantnew entry into petroleum
transportation.It was necessary for Rockefellerto react quickly if Stan-
dard was to maintainits monopoly position.
In December 1879, Standardpaid the Erie Railroad$50,000 to use its
right-of-wayto construct a pipeline from Bradfordto its Bayonne refin-
ery.9 At the same time Standardran a pipeline south from Bradford
along a branchof the PennsylvaniaRailroad.9'By 1881 Standardowned
pipelines that moved crude from the Oil Regions to Cleveland, to Pitts-
burgh, and to the New York harbor.92Once Rockefeller made these in-
vestments, he knew that a potentialpipelineentrantwould be faced with
the prospect of postentry transportationrates falling not to the marginal
cost of railroadtransportation,but to the lower marginalcost of pipeline
transportation.Therefore, once these pipelines were in place additional
pipeline entry was effectively deterred.
Standardwon the race to constructthe first long-distancepipeline net-
work because it had a fundamentaladvantageover other potential pipe-
line entrants. Because Standardcontrolledthe preexistingrail lines, any
other potential pipeline entrant knew that if it entered the market, the
price of petroleumtransportationwould fall to the marginalcost of rail-
road transportation.However, if Standardentered first, the price of pe-
troleum transportationwould remainat the monopoly level. As a result
of this asymmetry, Standardpossessed an added incentive over other
potential pipeline entrantsto enter first, and the race between Standard

90Rockefellerassured the Erie that its oil trafficwould be protected; 1 Nevins, supra
note 16, at 358-59.
9' Williamson& Daum, supra note 1, at 449.
92 2 Tarbell,supra note 6, at 12; Williamson& Daum, supra note 1, at 448-49. By 1883
Standardalso completeda pipelineto Philadelphia,includinga branchto Baltimore.See
Johnson, supra note 78, at 104.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
36 THE JOURNAL OF LAW AND ECONOMICS

and the other potentialpipelineentrantsdid not dissipateall of Standard's

transportationprofits.
Rockefeller also had to adopt a commitmentstrategyto drive out any
new pipelines that entered after Standard.This was necessary because
the prospect that Standardwould bringpipeline entrantsinto the trans-
portation conspiracy, as was done in the case of the Tidewater, would
encourage additionalpipeline entry and dissipate the cartel profits. An
example of this predatorycommitmentstrategy occurred in 1881 when
the Buffalo and Rock City PipelineCompanybuilt a crude pipeline from
the Oil Regions to Buffalo.93Buffalo was a desirable destinationfor an
independentpipeline because from Buffalothe crude or refinedcould be
moved by barges along the Erie Canaland then down the Hudson River
to New York, thus completelyavoidingStandard'sassociated transporta-
tion network. As expected, on completionof the Buffalo and Rock City
Pipeline, independentrefinersentered in Buffalo. In response, Standard
constructed a parallel pipeline along the same route which it used to
drive down transportationrates, destroying the Buffalo and Rock City
Pipeline's business.94The Buffalo and Rock City Pipeline then sold out
to Standard,whereupon Standardimmediatelyraised the rates from 10
cents to 25 cents per barreland tore up its own pipeline.9
After Standard'spipeline network was in place, Standardthen con-
trolled a feasible transportationalternative,and the railroads'bargaining
power (alongwith their shareof profits)decreased substantially.This can
be seen if we once again compare the rail rates for petroleumproducts
with the rates chargedto transportgrain and coal presented in Table 3.
In 1880, after the entry of the Tidewater,petroleumrail rates decreased
substantially, both absolutely and relatively. And in 1884, after Stan-
dard's pipeline network was in place and the final Tidewateragreement
with Standardwas reached, petroleumrates decreased even further.
At this point in time, Standardreorganizedits refiningcapacity, shut-
ting down some inlandlocations such as Pittsburghand adding capacity
on the coast to supply the export market,and the Standard-ownedpipe-
line transportationnetworkcarriedthe majorityof crude oil shipments."

93 The pipeline was constructed under the New York State Free Pipeline Act of 1878
which facilitated the acquisitions of rights-of-way in New York. See Williamson & Daum,
supra note 1, at 453. It was completed in spite of the fact that Standard and the Erie
attempted to prevent its construction (Derrick's, supra note 40, at 330; testimony of C. B.
Matthews, Trusts, supra note 1, at 424-25).
94 Testimony of C. B. Matthews, Trusts, supra note 1, at 425.
95 Id. After Standard gained control of the Buffalo and Rock City Pipeline it also acquired
the independent refiners in Buffalo (Hidy & Hidy, supra note 33, at 100).
9 In 1885 the seaboard refineries handled two-thirds of Standard's crude, most of it
delivered by Standard's pipelines (id. at 101). Since Standard now shipped most of its crude

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 37

However, althoughrailroadpetroleum shipmentswere relatively small,

the railroadscould dramaticallyincreasetheir shipmentsby offeringinde-
pendent refiners (includingnew entrants)lower rates. Therefore, Stan-
dard continued to provide the railroadswith a sufficient profit share in
the form of supplementaldirect payments made directly to the railroads
(and Standardcontinuedto maintaina dominantshare of refiningin order
effectively to police the railroads).For example, the August 1884agree-
ment with the Pennsylvania included a provision by which Standard
agreed to pay the Pennsylvaniafor any shortfallbetween actual petro-
leum shipments and their guaranteed26 percent.97In addition, in early
1884 Standardgave the Erie a direct subsidy of 27/8cents per barrel on
all crude that Standardshippedto its coastal refineries,whether by pipe-
line or rail. Since by 1887 the three railroadstogether carried only 24
percent of the crude shippedto the refiningcenters, these side agreements
entailed very large direct paymentsby Standardto the railroads.98"

D. Discovery of New Fields

Standard's control of the transportationof petroleum was complete
after the constructionof its pipeline network and the negotiationof the
final collusive marketsharingand price-settingagreementswith the rail-
roads and the Tidewaterin 1884.Majornew threatsto Standard'scontrol
of the petroleumindustrywould not come untilnew oil fields were discov-
ered. In 1891the McDonaldoil fieldwas discoveredon the Pennsylvania-
West Virginiaborder. This initiallydid not disturbStandard'scontrol of
transportation.The Producers'Oil Companyin the McDonaldfield built
a pipeline to a rail depot near Pittsburgh,where the crude was then
shippedon the Pennsylvaniato be refinedin the east for export.99Because
of highrailrates, the Producers'Oil Companyestablisheda new company
with a group of independentrefinersin the Oil Regions and extended the
pipeline from the McDonaldfield all the way to the Oil Regions, where
the crude was refined. However, because the refined had to be shipped
on Standard-controlledrailroadsfrom the Oil Regions to the East Coast,
the problemof disadvantageoustransportationrates remained.This moti-

by pipeline to be refinedat the coast for export and was not shippinga large volume of
either crude or refinedon the railroads,establishmentof the ICCin 1887(whichmeantthat
the railroadsnow had to chargethe same rate to all shippersand Standardcould no longer
receive any rebates)was not a problemfor Standard.(See answersto interrogatoriespro-
vided by John D. Rockefeller,IndustrialCommission,supra note 72, at 795.)
97 2 Tarbell,supra note 6, at 310-13.
98 Johnson, supra note 78, at 116, 270, n.71; Hidy & Hidy, supra note 33, at 726, n.18.
9 Williamson& Daum, supra note 1, at 569.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
38 THE JOURNAL OF LAW AND ECONOMICS

vated the group to form a company with Lewis Emery, Jr., called the
United States Pipe Line. The United States Pipe Line constructed two
parallelpipelines from the Oil Regions to the coast, a crude pipeline and
the first refinedpipeline.'" These lines were fought vigorously by Stan-
dard and the railroads,but were eventually completed in 1893.10' This
led to low transportationrates and substantialoperatinglosses for the
pipeline company in 1893and 1894.102In 1895,the three largestindepen-
dent refinerssold their plants and their shares in the pipeline company
to Standard.103 The remaininggroup of refinersestablished the Pure Oil
Company, which by 1900accountedfor only about4 percentof illuminat-
ing oil deliveries to the coast.10"

100 McGee, supra note 5, devotes a great deal of space to Emery, claimingthat Emery
continuallyenteredandprofitablysold his refiningcompaniesto Standard.On the contrary,
Emery'sbehaviorconfirmsourinterpretationof events. Emeryoriginallyenteredinto refin-
ing with the Octave Oil Companyand Refineryin the Oil Regions in 1870which he then
shutdownin 1875and sold to Standardin 1876.Emery'sexplanationfor the shutdownand
sale is that he and other "refinerswere pushedinto consolidationbecause of the rail rate
'preferences'given to Standardin the SouthImprovementCompanyprogramand the suc-
ceeding pool agreementof 1874" (quoted in McGee, supra note 5, at 145, n.19). McGee
rejects Emery's explanationof events because, he notes, Emerytestifiedthat Octave and
other small refiners received railroadrebates. However, McGee fails to distinguishthe
period Emery's testimony regardingrebates is referringto. Emery's testimony refers to
"the very early history of the Octave Oil Company"(McGee, supra, at 145, n.22) when
rebates were available to all refiners. It was after Standardestablishedand enforced a
railroadcartelthat smallrefinerssuch as Octavedid not receive rebates,and, as accurately
describedby Emery, Octave was squeezedby differentiallyhigh railroadrates and forced
to sell out to Standard.
Emery reenteredrefiningin 1880because the PennsylvaniaRailroadinvitednew refiners
to enteralongthe Pennsylvania'sline by guaranteeinglow freightratesto all refiners(Emery
testimony,Trusts,supra note 1, at 235-36). The Pennsylvania'soffer was madeduringthe
brief competitive period when the Pennsylvaniaviolated the tentative Tidewaterpooling
agreementand offered open, equal, and greatly reduced rates to all shippers. The final
Tidewaterpoolingagreementin 1884resultedin independentrefinersfacinga substantially
highertransportationrate(2 Tarbell,supranote 6, at 166.)Emeryhungon at these disadvan-
tageous rates for a while, but eventuallysold out to Standardin 1887. When testifyingin
1888, Emery stated that his Philadelphiarefinery"did not prove to be a profitableinvest-
ment" (Emerytestimony,Trusts,supra,at 236).Therefore,these two purchasesof Emery's
plants, first in 1876 and then again in 1887, are not examples, as McGee claims, of an
independentrefinerrepeatedlybribedby Standardto exit the industryin returnfor a share
of the monopoly profits. Ratherthan collecting a lucrativepaymenttwice, the evidence
indicates that Emery was squeezed out of refiningtwice, before learningthat successful
entry requireda way to evade Standard'stransportationsystem.
101To disruptconstructionof the lines, Standardandthe railroadsacquiredthe pipelines'
rights-of-way,physicallydisruptedthe lines at rail crossings, acquiredshares in the com-
pany, and filed legal challenges.The pipelineseventuallyconnectedto the Jersey Central
Railroad,a non-Standard-affiliatedrailline (Emerytestimony,IndustrialCommission,supra
note 72, at 650-53; Johnson,supra note 78, at 174-75).
102 Lee testimonyin United States v. StandardOil of New Jersey (1906),transcriptVI,
3164-65, from Williamson& Daum, supra note 1, at 575.
103 Williamson& Daum, supra note 1, at 576.
'04 Id. at 580.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 39

While Standardmaintainedits control of transportationfrom the Oil

Regions and from the McDonaldfield, it had no advantagein controlling
transportationfrom new oil fields that did not rely on the Standardtrans-
portationnetwork. As a result, many independentrefinersfreely entered
these new fields. For example, afterthe Lima Field in northwesternOhio
near Toledo was developed in 1885, many new independentrefinersen-
tered and grew, the most importantof which was Sun Oil. The refined
productsfrom Lima crudewere transportedalongraillinks which already
existed and were independentof Standardrunningfrom the Toledo area
to Chicago, Cleveland, and Buffalo.1?5Similarly,after the importantdis-
covery of the Texas oil fields in 1901,additionalmajorindependentrefin-
ers entered and grew, includingGulf Oil and the Texas Company(Tex-
aco). Gulf and Texas transportedtheir crude on their own pipelines to
their refineriesat Port Arthur,Texas, on the Gulf of Mexico. From these
refinerieskerosene could be transporteddirectly to Europe by ship. In
addition,a pipeline system developed by a groupof independentpipeline
companies acted as a common carrierand transportedTexas crude for
smaller refiners.106A great many small new refiners also successively
entered after new discoveries were made in the Mid-ContinentField in
Kansas and Oklahomain 1905andthe Illinoisfieldin 1906.107Independent
refinerssurvived and prosperedon crude from these fields because there
was no role in any of these new fields for Standardto act as an evener
of a petroleumtransportationcartel.
It is importantto recognize that even after the discovery of these new
fields and Standard'sloss of control of the U.S. refiningindustry, Stan-
dardcontinuedto maintainits monopsonypower in the purchaseof crude
from fields in the Oil Regions. This is because Standardcontinued to
retain control of transportationfrom the Oil Regions with its pipeline
system. Only when somewhat effective ICC regulationof pipelines was
established in 1914 did Standardfinally begin to lose its monopsonistic
power over the purchaseof Oil Regions crude.108
105
RalphLouis Andreano,The Emergenceof New Competitionin the AmericanPetro-
leum Industrybefore 1911,at 29, 45 (Ph.D. Dissertation,NorthwesternUniversity 1960).
106 Williamsonet al., supra note 4, at 75-78.
107 Id. at 16, 24, 90.
108 Althoughpassageof the Hepburnamendmentin 1906made interstatepipelinescom-
mon carriers,the fact that pipelinecompanieswere organizedby state andtransportedonly
theirown oil led some state courtsto concludethatthe pipelineswere not commoncarriers.
As a result,manypipelinecompanies(includingStandard)viewedthemselvesas proprietary
and, therefore, not obligatedto file tariffs with the ICC. In 1914the SupremeCourt dis-
agreed (Pipe Line Cases, 234 U.S. 548 (1914)).However, even after this decision and the
priordivestitureand breakupof Standard'spipelinesystem in 1911,Standardcontinuedto
limit access to theirtransportationnetworkfromthe Oil Regionsby, for example,requiring
large minimumshipments.InterstateCommerceCommissionregulationattemptedto pre-
vent such exclusion, but was not always successful.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
40 THEJOURNALOF LAWAND ECONOMICS

VI. EFFICIENCYEXPLANATIONS
FORROCKEFELLER'S
SUCCESS

Many authors have argued that Rockefeller's managerialcost advan-

tages, such as his ability to take advantage of economies of scale and
more closely coordinateinput and output flows to maintainhigh rates of
capacity utilization, can explain the growth of StandardOil during the
1870s.109The rapiddecline of crude and refinedprices duringthe period
of Standard's consolidation, presented in Figure 3, appears consistent
with such efficiency explanations.Furtherconfirmingevidence is the fact
that output nearly quadrupledduringthe 1870s.110The large expansion
in petroleumsupply duringthe 1870s,however, can be attributedprimar-
ily to the discovery of new fields withinthe Oil Regions and to the more
extensive developmentof existing fields, not to any technicalor manage-
rial improvementsin refiningefficiency instituted by Standard."' But,
whatever the reason for the shift in supply, the decrease in price and
increase in quantityduringthe 1870sis not inconsistentwith a simultane-
ous growth of marketpower.
The idea that the success of Rockefellerduringthe 1870scan be attrib-
uted to his superiorefficiencyis contradictedby the timingand behavioral
evidence associated with the growth of StandardOil. There is no evi-
dence of any significanttechnologicalchange that occurredin refiningin
late 1871, the point in time when Standardacquired essentially all the
refineriesin Cleveland.Moreover, superiormanagementtechniques can-
not explain the rapidityand completeness of Rockefeller's acquisitions.
On the contrary,Rockefeller'sacquisitionsin Clevelandoccurredimme-
diately after the establishmentof the South ImprovementCompanyand
were facilitated by officers of the Company and railroadofficials, who
convinced independentrefiners that they would henceforth face disad-
vantageous rail rates.
Standarddid take advantageof efficienciesassociated with transporta-
tion economies of scale by guaranteeingrailroadslarger shipments in
return for lower rates. However, it was not necessary for a refiner to

'09For example, 1 Nevins, supra note 16, at 154-58, 175, and, more recently, Alfred
D. Chandler, Jr., Scale and Scope: The Dynamics of Industrial Capitalism 24-25
(1990).
110From 1870 to 1879 yearly productionrose from 5,261 to 19,914 thousand barrels
(Leven,supranote70, at 40).
"I The numberof producingwells increasedfrom 4,363 in 1870to 15,521in 1879. Id.
Crude producersexpanded productionin the face of falling prices because the Rule of
Capturemeantthatcrudeoil propertyrightswere establishedonly whena producerbrought
the oil to the surface. Therefore,once more than one producertappeda field, it paid all
the producersoperatingin the field to increasethe numberof wells and pumpoil from the
field as rapidlyas possible.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 41

own all the refineriesin Cleveland, much less the entire United States,
to obtain such transportationefficiencies. Before Standard'sconsolida-
tion in Cleveland, other Clevelandrefinerswere obtainingrebates similar
to Standardby makingtransportationcommitments.If additionalecono-
mies of scale were possible, refinerscould have combinedtheir shipments
without merging.112
We also have seen that independentrefinersfreely enteredand shipped
their productin competitionwith Standardwhen new fields were discov-
ered. Such entry also occurredwhenever the transportationcartel broke
down, such as duringthe EmpireWar in 1876-77 or duringthe competi-
tive episodes of 1880-83. If Standard'srate advantagewas purely cost-
based, Rockefeller would not have had actively to prevent the railroads
from offeringdiscounts from agreed-onrailroadrates to independentre-
finers. Standard's ability to take advantage of superior transportation
efficiency was not what drove independentrefiners of the Oil Regions
crude out of the marketnor what kept them out.
Lester Telser presents an alternativetransportationefficiency explana-
tion for Standard'ssuccess that is based not on economies of scale in
schedulingor other large shipmentefficiencies, but on naturalmonopoly
conditions present in the railroadindustry.113 Telser argues that because
of the railroads' high fixed costs and relatively low marginalcosts, a
stable competitive equilibriumwas not feasible.114 Telser claims that
Standardsolved this naturalmonopolyproblemby eliminating"destruc-
tive competition" among the railroadsand that the resultingstability in
the transportationindustryled to lower consumerprices.
Telser's efficiency theory is similar in some fundamentalrespects to
our theory of railroadcartelization.As we described above, "ruinously
competitive" episodes existed in petroleumtransportationbefore 1872,
and Standard'senforcementof a railroadcarteleliminatedthese episodes
and stabilizedrailroadrates. However, our analysis differsfrom Telser's
analysis in some major respects. In particular,Telser does not analyze
the process by which Standardcreated and enforced a petroleumtrans-

112 For example, before Standard'sacquisitionstwo other Clevelandrefinerstogether

with Standardhad a joint agreementwith the Erie Railroadwhere they received significant
rebatesfor guaranteedshipments.See ChesterDestler, The StandardOil, Childof the Erie
Ring, 1868-1872,33 Miss. Valley Hist. Rev. 89, 104-6 (1946).
113 Lester Telser, A Theory of EfficientCooperationand Competition36-41 (1987).
114 While a
competitiveequilibriumdid not exist, some more complex dynamicequilib-
riummay have existed. The three railroadsthat initiallyenteredthe petroleumtransporta-
tion industry presumablybelieved that rates would fluctuateover time (as the industry
alternatedbetween rate wars and implicitor explicitrateagreements)and that, on average,
prices would be sufficientto cover the initialinvestmentin the railbed and tracksthat was
necessary to tie their existing rail systems to the Oil Regions.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
42 THE JOURNAL OF LAW AND ECONOMICS

portationcartel, and he discountsthe marketpower createdby Standard,

ignoringcompletely Standard'smonopsony of crude productionand in-
correctly claimingthat the efficiency associated with the creation of the
StandardOil Trust led to lower refinermargins.
Telser supports his theory by regressingover the 1874-95 period the
refined price against crude price, time, and time-squaredvariables, in
additionto a dummyvariableentered interactivelyfor the establishment
of the StandardOil Trust in 1882.115 This regressionleads Telser to con-
clude that, holdingthe crude price constant, the refinedprice was signifi-
cantly lower after the formationof the StandardOil Trust than it would
have been absent the Trust.
The assumptionthat Standardhad no effect on the crude price which
underliesthis empiricalanalysis makes no economic sense. Telser justi-
fies this assumptionby assertingthat "the oil trust did not then engage
in crude production to any significantextent, and producers of crude
could and did sell their product in a competitive internationalmarket,
especially to European importers."116 However, the fact that Standard
did not produceany significantcrudedid not preventStandardfrom exer-
cising monopsonypower in the crude marketthroughits control of trans-
portation.And this monopsonisticpower was not limitedby the fact that
some very small amount of crude was sold directly to Europe since all
petroleum, whether exported as kerosene or as crude, had to be trans-
ported to the coast on railroadsor pipelinescontrolledby Standard.Stan-
dard's controlof petroleumtransportationclearlypermittedit to exercise
monopsony power in the crude market.
In addition,Telser's empiricalanalysis of Standard'seffect on refining
margins uses an arbitrarymodel of what the trend in the refined price
relative to the crude price would have been absent the creation of the
Trust. Looking at Figure 3, it does not appearthat anything significant
occurs in 1882. However, because the refinerymargin,that is, the differ-
ence between the refinedprice and the crude price, generallyfalls over
the pre-Trustperiod at a decreasing rate, it is not surprisingthat the
regression of the refinedprice against the crude price finds a significant
negative time trend and a positive time-squaredtrend over the pre-Trust
period. Using this as the benchmarkof what would have occurredin the
post-1882periodif the Trusthad not been establishedinvolves extrapola-

"5 Telser's choice of 1882as the key dateis incorrect.Standardcontrolledtransportation

muchearlier,beginningits policingof the railroadsin 1874.By 1879,when Standard'sshare
of refiningwas 90 percent, its controlof transportationwas unquestioned.Whatoccurred
in 1882was merely a changein the legal form of the Standardorganization.
"116Telser, supra note 113, at 38.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 43

tion of the positive effect of the estimated quadraticterm from the pre-
Trust period into the post-Trustperiod. As we move furtherout in time,
the positive quadratictime term increasinglydominatesthe negative sec-
ular time term, with the result that the predicted "but for" refinedprice
increases dramaticallyover time. One cannot infer solely from the fact
that the refiningmargindeclined in the earlierperiod at a decreasingrate
that, without the establishmentof the StandardTrust, this trend in the
refining margin would have reversed itself and increased dramatically.
Evidence of the inappropriatenessof this extrapolationis the fact that
the prices and refinerymarginspredicted for the end of the estimation
period in 1895are substantiallyhigherthan the actualprices and margins
that existed before the Trustwas established,presumablya point in time
when the inefficiencies of ruinous competition that the Trust is alleged
to have later corrected were present.117
If one wished to use the movement of crude and refined prices to
determine whether Standardhad any monopoly-monopsonyimpact on
the industry, it would make economic sense, given the fact that crude
and refinedprices are decliningdramaticallyover time, to look not at the
refinerymarginbut at the ratio of the refinerymarginto the refinedprice.
This ratio increases dramaticallyin 1873 and generally remains at this
higher level for the following 2 decades. In particular,the ratio moves
from an average level of 61 percent during 1869-72 to an average of 75
percent during 1873-93, excluding the sharp drop in the ratio that oc-
curredduring 1876-77 (or to 74 percent if we include 1876-77). In either
case the null hypothesis that the means are equal in the earlierand later
periodscan be rejectedat the .01 level. Moreover,the increase in refinery
profitsreflectedin this ratioof the refinerymarginto the refinedprice that
occurs after 1872understatesRockefeller'smonopolisticimpact because
marginaltransportationand refiningcosts are excluded from the calcula-
tion of the ratio and decrease dramaticallyover the period.

VII. CONCLUSION
The most convincing evidence that Rockefellercreated marketpower
in the petroleumindustryby cartelizingtransportationis not the overall
patternof crude and refinedprices, but the detailed behavioralevidence
we have presented. The evidence includes the timing and completeness
of Standard'sinitialconsolidationin Clevelandafter the formationof the

The predictedrefinedprice and impliedrefinerymarginin 1895absent the Trust is

"117
more thantwo-and-a-halftimes the actuallevel of the price and marginat the startingpoint
of Telser's analysis in 1874and about four times the actual level of the price and margin
in 1881,immediatelybefore the Trustwas established.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
44 THEJOURNALOF LAWAND ECONOMICS

South ImprovementCompany, Standard'suse of preferentialrail rates

to acquire a dominantposition in the remainderof the refiningindustry,
Standard's vigorous opposition to the entry of long-distance pipelines
and then vertical integrationinto long-distancepipeline transportation,
Standard'sloss of dominanceof refiningonly when new fields were dis-
covered where it did not controltransportation,and, perhapsmost impor-
tant, Standard'svery effective cartel policing behaviorover more than 2
decades.
Standard'scartelpolicingbehaviorincludedthe stabilizationof individ-
ual railroadmarketshares and rail rates and the punishmentof railroads
that deviated from this cartel agreement. An obvious example of such
cartel policing was the shift by Standardof all its productionaway from
Pittsburghand Philadelphiato its refineriesin Clevelandand New York
to punish the PennsylvaniaRailroadduring the Empire Transportation
CompanyRate Warof 1876-77. This exampleillustrateswhat we believe
was Rockefeller's unique talent, namely, his ability to negotiate and en-
force a collusive arrangement,especially his willingnessto bear the short-
term costs necessary to punish any railroadthat deviatedfrom the collu-
sive transportationagreement.WhileRockefellercertainlywas a talented
managerand organizer, it was these negotiationand commitmentskills
which enabled him to monopolizethe petroleumindustry.
Finally, our analysis of the growth and maintenanceof StandardOil's
market power sheds light on how a vertical arrangementcan have an
anticompetitiveeffect. Contraryto establisheddoctrinethe vertical rela-
tionship between Standardand the railroadsdid not increase "barriers
to entry" into the petroleumindustryby requiringthat entry occur on
two stages."118 The evidence is clear that, whenever entry occurred into
transportation,as in the B&O-ConduitPipelinecase, the TidewaterPipe-
line case, or the Buffaloand Rock City Pipelinecase, competingrefiners
were availablefor the new supplierof transportationservices. Although
the vertical connection between Standardand the railroads did make
unilateral entry into refiningmore difficult, the binding constraint was
entry into transportation,and the verticalrelationshipdid not make entry
into transportationmore difficult.The sole anticompetitiveeffect of the
vertical relationshipwas in facilitatingthe horizontalconspiracy, initially

"8 This is the anticompetitivemechanismemphasizedin the U.S. Departmentof Justice

MergerGuidelineswhen evaluatingverticalmergers(U.S. Departmentof Justice Merger
Guidelines1984,? 4.2). The 1992revised MergerGuidelines,jointly issued by the Depart-
ment of Justice and the FederalTrade Commission,do not include a section on vertical
mergers and the Statementaccompanyingrelease of the Guidelinesnotes that "Neither
agency has changedits policy with respect to non-horizontalmergers"(U.S. Department
of Justice release, April2, 1992).

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 45

among the railroadsserving the Oil Regions and, later, among the rail-
roads and the Standardpipeline network.
Therefore, although the StandardOil case vividly illustrates how a
vertical relationship can facilitate the creation of horizontal market
power, the vertical relationshipitself is not the fundamentalaspect of the
arrangementthat shouldbe of concernto antitrustregulations.Standard's
use of its vertical position in refiningto police the railroadseffectively
did stabilize the petroleumtransportationcartel. However, the explicit
horizontalconspiracythat was undertakenby the railroadswith the assis-
tance of Standard, a conspiracy which jointly set rail rates and fixed
individualrailroadmarketshares, was necessary for any anticompetitive
effect. Such horizontalcollusive behaviorclearly is anticompetitive,and
would be anticompetitiveeven if there were no vertical connection be-
tween Standardand the railroads(for example, if the collusive arrange-
ment were enforced by a clearinghousearrangementestablished by the
railroads).Hence, while we have carefullydocumenteda case of "raising
rivals' costs," our analysis providesno supportfor a new antitrustpolicy
which would condemn a vertical relationshipwithout the presence of a
horizontalconspiracy.

BIBLIOGRAPHY

Adams, CharlesFrancis, Jr. Railroads:TheirOriginand Problems. New York:

G. P. Putnam'sSons, 1878.
Andreano, Ralph Louis. The Emergence of New Competitionin the American
PetroleumIndustrybefore 1911. Ph.D. dissertation,NorthwesternUniversity,
1960.
Bryce,Lloyd."TheRailwayProblem:
I. TheLegislativeSolution."NorthAmer-
ican Review 164 (1897):327-38.
Burgess, GeorgeH., and Kennedy,Miles C. CentennialHistoryof the Pennsylva-
nia Railroad Company,1846-1946. Philadelphia:PennsylvaniaRailroadCom-
pany,1949.
Chandler, Alfred D. Scale and Scope: The Dynamics of Industrial Capitalism.
Mass.:Belknap/Harvard
Cambridge, UniversityPress, 1990.
The Derrick's Hand-Book of Petroleum: A Complete Chronological and Statisti-
cal Review of Petroleum Developments from 1859 to 1898. Oil City, Pa.: Der-
rickPublishing Co., 1898.
Destler,Chester."TheStandard
Oil,Childof theErieRing,1868-1872."Missis-
sippi ValleyHistoricalReview 33 (June 1946):89-114.
Dodd, S. C. T. Combinations: Their Uses and Abuses with a History of the
StandardOil Trust.New York:George & F. Nesbitt Co., 1888.
Frey, Robert L., ed. Railroads in the Nineteenth Century. New York: Bruccoli
ClarkLeynau,1988.
Friedman,Milton, and Schwartz,Anna J. MonetaryTrendsin the United States

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
46 THE JOURNAL OF LAW AND ECONOMICS

and the United Kingdom: Their Relation to Income, Prices, and Interest Rates,
1865-1975. Chicago:Universityof ChicagoPress, 1982.
Henry, J. T. The Early and Later History of Petroleum with Authentic Facts in
Regard to Its Development in Western Pennsylvania. 2 Vols. Philadelphia: Jas.
B. Rodgers Co., 1873;New York: Burt Franklin,1873.
Hidy, Ralph W., and Hidy, Muriel E. Pioneering in Big Business, 1882-1911:
History of Standard Oil Company (New Jersey). New York: Harper & Broth-
ers, 1955.
Hilton, George W. "The Consistencyof the InterstateCommerceAct." Journal
of Law and Economics 9 (October 1966): 87-113.
Hovenkamp, Herbert. "RegulatoryConflict in the Gilded Age: Federalism and
the RailroadProblem." YaleLaw Journal97 (1988): 1017-72.
Hubbard, Freeman H. Encyclopedia of North American Railroading: 150 Years
ofRailroading in the United States and Canada. New York: McGraw-Hill, 1981.
Industrial Commission. Preliminary Report on Trusts and Industrial Combina-
tions. Vol. 1. Washington,D.C.: U.S. GovernmentPrintingOffice, 1900.
Interstate Commerce Commission. Railways in the United States in 1902. Pt. II.
Washington,D.C.: U.S. GovernmentPrintingOffice, 1903.
Johnson, Arthur Menzies. The Development of American Petroleum Pipelines:
A Study in Private Enterprise and Public Policy, 1862-1906. Ithaca, N.Y.:
CornellUniversity Press, 1956.
Krattenmaker,Thomas G., and Salop, Steven C. "AnticompetitiveExclusion:
Raising Rivals' Costs to Achieve Power over Price." Yale Law Journal 96
(December 1986):209-93.
Krattenmaker,ThomasG., and Salop, Steven C. "Competitionand Cooperation
in the Market for Exclusionary Rights." American Economic Review Papers
and Proceedings 76 (1986): 109-13.
Leven, David D. Petroleum Encyclopedia: "Done in Oil." New York: Ranger,
1941.
McGee, John S. "PredatoryPrice Cutting:The StandardOil (N.J.) Case." Jour-
nal of Law and Economics 1 (1958): 137-69.
Maybee, Rolland Harper. Railroad Competition and the Oil Trade, 1855-1873.
MountPleasant, Mich.: Extension, 1940.
Montague, Gilbert Holland. The Rise and Progress of the Standard Oil Company.
New York: Harper& Brothers, 1903.
Nevins, Allan. Study in Power: John D. Rockefeller, Industrialist and Philanthro-
pist. Vols. 1 and 2. New York: CharlesScribner'sSons, 1953.
New York Assembly. Proceedings of the Special Committee on Railroads, Ap-
pointed under a Resolution of the Assembly to Investigate Alleged Abuses in
the Management of Railroads Chartered by the State of New York. Vols. 1-8.
New York: Evening Post Steam Press, 1879-1880.
Ordover,JanuszA.; Saloner,Garth;and Salop, Steven C. "EquilibriumVertical
Foreclosure." AmericanEconomicReview 80 (March1990):127-42.
Petroleum Producers' Union. A History of the Rise and Fall of the South Im-
provementCompany.Reportof the ExecutiveCommittee.Lancaster,Pa.: Wy-
lie & Griest, 1872.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions
 MONOPOLIZATION 47

Poor, Henry V. Manual of the Railroads of the United States. Series nos. 4, 6,
7. New York: H. V. & H. W. Poor, 1871, 1873, 1874.
Railroad Gazette. Vols. 4-9. New York: S. WrightDunningand M. N. Forney,
1872-1877.
Salop, Steven C., and Scheffman,David T. "RaisingRivals' Costs." American
Economic Review 73 (May 1983): 267-71.
Schlegel, Marvin W. Ruler of the Reading: The Life of Franklin B. Gowen, 1836-
1889. Ph.D. dissertation, Columbia University; Archives Publishing Co. of
Pennsylvania, 1947.
Stickney, A. B. TheRailroadProblem. St. Paul, Minn.:D. D. MerrillCo., 1891.
Tarbell, Ida M. The History of the Standard Oil Company. Vols. 1 and 2. New
York: MacMillanCompany, 1904.
Telser, Lester G. A Theory of Efficient Cooperation and Competition. New York:
CambridgeUniversity Press, 1987.
U.S. Bureau of Corporations. Report of the Commissioner of Corporations on
the Petroleum Industry. Pts. 1 and 2. Washington,D.C.: U.S. Government
PrintingOffice, 1907.
U.S. House of Representatives,Committeeon Manufactures.Report on Investi-
gation of Trusts.House ReportNo. 3112. Washington,D.C.: U.S. Government
PrintingOffice, 1888.
Wait, JamesJ. "The RailwayProblem:II. A MercantileView." North American
Review 164 (1897):338-48.
Williamson,Harold F.; Andreano,Ralph L.; Daum, ArnoldR.; and Klose, Gil-
bert C. The American Petroleum Industry: The Age of Energy, 1899-1959.
Evanston, Ill.: NorthwesternUniversity Press, 1963.
Williamson, Harold F., and Daum, Arnold R. The American Petroleum Industry:
The Age of Illumination, 1859-1899. Evanston, Ill.: Northwestern University
Press, 1959.
Yergin, Daniel. The Prize: The Epic Quest for Oil, Money and Power. New York:
Simon & Schuster, 1991.

This content downloaded from 209.175.73.10 on Tue, 05 Jan 2016 07:12:41 UTC
All use subject to JSTOR Terms and Conditions