Economic History Association
Cartels and Regulation: Late Nineteenth-Century Railroad Collusion and the Creation of the
Interstate Commerce Commission
Author(s): Thomas S. Ulen
Source: The Journal of Economic History, Vol. 40, No. 1, The Tasks of Economic History
(Mar., 1980), pp. 179-181
Published by: Cambridge University Press on behalf of the Economic History Association
Stable URL: http://www.jstor.org/stable/2120445
Accessed: 10/02/2010 16:42
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=cup.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit 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 [email protected].
Cambridge University Press and Economic History Association are collaborating with JSTOR to digitize,
preserve and extend access to The Journal of Economic History.
http://www.jstor.org
Summaries of Dissertations
179
that with the advent of "social innovation," entrepreneurship is channeled into controlling
the firm's social environment, with significant social and economic consequences. By incorporating "social innovation" into the Schumpeterian method, we at the same time
broaden and expand the usefulness of that method. Schumpeter's pessimistic prognosis of
the demise of capitalism due to a lack of entrepreneurial dynamism is, however, called
into question.
DAVID J. ST. CLAIR, California State University-Hayward
Cartelsand Regulation: Late Nineteenth-Century
Railroad Collusion and the Creationof the
Interstate CommerceCommission
This work reexamines the operation of the trunkline railroad cartel between Chicago
and the Atlantic coast from the mid- 1870s to the mid- 1890s in order to illuminate two important issues in economic history and theory.' First, a thorough study of the railroad cartel reveals that in the years before the creation of the Interstate Commerce Commission
and the passage of the Sherman Act, the collusion was far more successful at maintaining
near-monopoly profits for its members than the current theory of cartel behavior would
predict or the previous investigators of the cartel have found. This finding necessitated a
new theory of collusion and a test of its hypotheses against the relevant data in order to try
to ascertain what circumstances contribute to cartel stability. Second, if the railroad cartels
exhibited stability before 1887, then the now-accepted hypothesis that the Interstate Commerce Commission was created to bring stability to otherwise chaotic railroad collusion
must be reconsidered.
The Chicago Eastbound Dead Freight Pool was the formal organization that handled
collusion on flour, grain, and provisions after June 1879. The cartel was an elaborate market-share-assigning and rate-setting arrangement among the through lines between Chicago and all eastern ports north of Baltimore. Initially there were six members, but by
1884 there were nine, the additional roads having been encouraged to enter the industry
by the cartel's success. In addition to the familiar problems of price-shading to which any
cartel is prey, the Chicago Pool faced other complications, all of which should have increased the collusion's instability. The eastbound agreement covering dead freight was
only part of a far more extensive cartel which dealt with live freight from Chicago to the
coast, with both dead and live through-freight from midwestern commercial centers other
than Chicago, and with westbound shipments of all commodities from the coast to the
Midwest. In addition, there was the threat of competition from alternative transport
modes-the Great Lakes route or the Mississippi-coastwise route-and from entrants into
the rail transport market. The elaborate transport market over which this body attempted
to hold sway clearly presented myriad opportunities for cheating.
The cartelists recognized these problems and established internal policing mechanisms
with penalties for violating the implicit cartel contract. All of these separate commodity
and locational cartels were governed by the Joint Executive Committee, which was, in essence, a private government with an executive, a legislature, and a judiciary. The JEC had
five means of reducing the incentive to cheat. (1) The group compiled and published
weekly statistics on tonnage movements by members; these were verified both by station
'This dissertationwas completedat StanfordUniversityunderthe directionof Paul A. David.
180
Summaries of Dissertations
agents and by employees of the Chicago Board of Trade. (2) The JEC hired a Board of
Arbitrators to settle disputes and assign market shares. The men who served in this role
were very handsomely paid and well known nationally (for example, Thomas Cooley, David Wright, and Charles Francis Adams). (3) From March 11, 1881, the Commissioner of
the JEC, Albert Fink, had the power to match immediately any price cut discovered on
the floor of the Board of Trade. (4) Various economic sanctions were imposed on a discovered cheater. Good faith deposits were forfeited in whole or in part, and the commissioner
forbade any member to transfer freight with a cheater. (5) The most powerful inducement
to loyalty was the frequent but irregular attempts to equalize actual and allotted market
shares of freight. Independent agents of the cartel directed shippers from roads over their
allotments to those under their allotments. The traffic so diverted in the dead freight pool
was, by force of circumstance, largely flour and grain bound for Europe.
An examination of contemporary reports on the cartel's affairs in the business press and
in the JEC's memoranda indicates that during the 328 weeks between January 1880 and
April 1886, there were reports of cheating in fewer than one-third of those weeks. Cogent
arguments are offered that these reports under-report adherence to the cartel contract in
the business upturn and over-report it in the downturn. This finding-of considerable collusive stability because of internal enforcement measures on the part of the cartelprompted a new theory of collusive behavior and an empirical test of its hypotheses. In the
presence of internally effective punishment, cheating becomes a gamble: the probability of
being detected necessitates comparison of the expected cheating profits with expected loyalty profits. I assume that the firm takes the action with the higher expected profit. The
model developed assumes that the probability of being detected by other cartel members
is endogenous, depending on, inter alia, the amount of output over which a secret price
discount is offered. Using a stylized cyclical movement in the demand for the industry's
output, I show that each cartel member will find it more profitable to remain loyal when
demand is expanding and to cheat when demand is contracting.
The model is then tested against the newly gathered weekly data from the Chicago
Eastbound Dead Freight Pool for the period 1880 to 1886. A binary dependent variablethe probability of adherence-is defined from four independent contemporary journals,
the value of the variable being 1 if there were no reports of cheating in any of the journals,
and 0 otherwise. The estimation used the logistic function, in which the vector of independent variables included proportional deviations between allotted and actual market
shares for each cartel member, the total tonnage shipped east by the cartel in the previous
period, the state of the cartel in the previous week times the number of tons below allotments, a dummy variable to indicate whether lake navigation was open or closed, and the
absolute value of all deviations from allotments weighted by mean tonnage of flour and
grain over the last month. The estimation found all of those variables statistically significant with the most significant variables in determine the probability of cartel adherence
being the total tonnage shipped east by the cartel and the size of the transferable subset
(flour and grain) relative to deviations from allotments.
These results form the basis for a reassessment of the intent and effect of the first federal
regulatory agency. The record of the railroad collusion after 1887 is explained by the same
factors that accounted for success before regulation: the volume of flour and grain shipments from Chicago increased at a rapid rate from 1886, not declining until 1893, the
longest period of continuous increase in the demand for transport in the late nineteenth
century. Additionally, entry into the railroad business declined from 1887 to 1897, for reasons not associated with federal regulation, freeing the cartel from the disruptive effects of
new competitors. When the Interstate Commerce Act is reexamined in light of this new
knowledge, it is found to have been, in intent, against collusion's interests. Most important
in this regard was section 5, which made illegal the pooling of freight or revenues by independent railroads. The other clauses that have been held to have been crucial in guaranteeing cartel stability were either too vague, like the long haul-short haul clause, or al-
Discussion
181
ready standard industry practice, like the posting of all rate schedules and changes. The
ICC should have, on balance, reduced the cartel's success, but for its first ten years of existence, the agency was ineffectual.
THOMAS S. ULEN, Universityof Illinois
Discussion
It's harvest time again, and this year's crop of dissertations in economic history makes a
rich harvest indeed. Each of the five dissertations dares to tackle large and tough issues;
each offers a basic reinterpretation; and in each case, the analysis is firmly rooted in new
data found deep in the archives. My task is twofold. I shall offer extra advertising for these
fine products, and shall point out ways in which their analyses can be extended, confining
myself here to the theses by Lamoreaux, Ulen, and Mirowski.
Naomi Lamoreaux has enriched our understanding of the peculiarly pronounced surge
of industrial mergers between 1898 and 1902. To Ralph L. Nelson's earlier study of merger
waves she adds two cross-sectional tests and a theoretical twist.' The first test is the detailed study of a family of related industries, here the steel and paper sectors. These chapters (3, 4, and 6) are a model of clarity and logic. Industries that are characterized by marginal cost curves running "down-flat-up," such as tinplate, wire, and newsprint, had a
special reason to merge, because demand slumps would induce them to slash prices and
keep "running full." Their urge to merge was reinforced by instability in the 1890s. Before
1893 surging demand lured them into expanding capacity, and the crash thereafter forced
them to merge or be strangled by prices below average cost. Lamoreaux convincingly contrasts their fortunes with steel and paper industries that lack these features. She follows
through with a fine sequel chapter about the differential success of the steel and paper
mergers after 1901.
Lamoreaux's quantitative cross-section of the whole manufacturing sector (chap. 5)
builds a fair prima facie case for her main hypothesis, namely, that intra-industry merger
springs from that mixture of "down-flat-up" marginal cost curves and unstable demand.
Here she has an opportunity to give her hypothesis even stronger support. If she were to
use a regression technique appropriate to limited-value dependent variables, such as logit
regression, she would not have to confine herself to comparing merger with one independent variable at a time. With such regressions, she could test her full hypothesis, by entering
interaction terms that join industry fixed-cost shares with rates of output growth in the
1880s and 1890s. I think the final results will give even stronger support to her hypothesis.
She may in the future be able to extend her cross-sectional turn-of-the-century hypothesis to the task of explaining the curious time profile of American mergers. Ralph Nelson
argued that major improvements in capital markets just before the turn of the century explain that first wave. Lamoreaux only touches on this point, and has a chance to push further. She can also see if her hypothesis explains why there was less of a merger wave in the
late 1930s and 1940s, after the next great depression, than in the late 1920s.
Thomas Ulen, like St. Clair and Lamoreaux, delves into sellers' plots against buyers.
His study of railroad cartels in the 1880s and the coming of the ICC in 1887 makes four
offerings: (1) a call for more common sense in theorizing about cartels; (2) a new model,
new data, and new tests on the cyclical pattern of cartel adherence; (3) a readable chronol' Ralph L. Nelson, Merger Movements in American Industry (Princeton, 1959).