the demise of ocean shipping regulation: a study in the evolution of ...

1 Assistant Professor of Law, Cleveland State University; I welcome all feedback [email protected]. My thanks for feedback to Peter Carstensen and Spencer Weber Waller.Steve Calkins provided very useful comments on a slightly different version.

THE DEMISE OF OCEAN SHIPPINGREGULATION: A STUDY IN THE

EVOLUTION OF COMPETITION POLICYAND THE PREDICTIVE POWER OF

MICROECONOMICS By Chris Sagers1

JEL Codes: A11, B10, B20, B40, K20, K21, K23, L40, L93, N70



MICROECONOMICS Abstract:

This paper examines the recent deregulation of a major U.S. industry, theocean transport of goods by liner shipping. In addition to its general policysignificance, this recent experience sheds light on two particularly interestingphenomena in the history of U.S. competition policy: the use and abuse of abstracttheory in the making of public policy, and the historical evolution of the policy itself.

Over its 140 year history, ocean liner shipping has almost always enjoyed anantitrust exemption permitting price-fixing cartels of ocean carriers, premised on thebelief that problems of cost and capacity inherent in the trade can be resolved onlyby horizontal collusion. Now that that exemption has been whittled away byderegulatory efforts, the pre and post-deregulation evidence presents one of theworld’s rare opportunities for natural experiment on the behavior and effectivenessof collusive cartel pricing (rare because price-fixing is normally illegal and thereforedone in secret).

Because normal and effective competition really never existed prior to about1998, the normative foundation of the antitrust exemption was based almost entirelyon theoretically modeled economic arguments. Observing the industry’s behaviorunder deregulation is therefore a before-and-after opportunity to test the predictiveaccuracy of at least one body of economic argument. Interestingly, deregulatoryexperience so far suggests that liner markets can perform well under normal pricecompetition, contrary to long-standing claims from the industry and some academics.

The evolution of shipping policy is also part of a larger historiographicalstory. On the one hand, it mirrors very closely the overall evolution of competitionpolicy itself. So much is now different than it was in 1916, when the United Statesadopted its first shipping policy – markets are different, industry is different,microeconomics is different, and both the law and the law’s esteem for the views ofacademic economists are different. As it happens the nature and prominence ofstatutory displacements of antitrust in favor of regulation is also very different, andas will be seen the story of the shipping exemption reflects the larger story ofgovernment efforts to cope with perceived of the problems of industrial organization.

2 Arthur Alan Leff, Economic Analysis of Law: Some Realism About Nominalism, 60 VA.L. REV. 451, 477 (1974). Cf. Guido Calabresi, The Pointlessness of Pareto: Carrying Coase Further,100 YALE L. J. 1211, 1231 n.60 (1991) (“I do not mean to suggest that serious economists are likelyto be misled into thinking [public policy] is easy. But the economists’ ‘groupies’ – popularizers, andeven politicians who think they understand economics – may well be misled.”).

3 See HERBERT HOVENKAMP, ENTERPRISE AND AMERICAN LAW, 1836-1937, at 300-04 (1991)[hereinafter “HOVENKAMP, ENTERPRISE”].



MICROECONOMICS

To suggest . . . that [one] look at everything at once or even seriatim, would be vainand foolish advice. Counseling humility, of course, suffers from no suchdrawbacks.

– Arthur Alan Leff 2

This paper, predominantly a study of a recent change in federal policy, insome important sense cannot help but also be about a deeper phenomenon in theAmerican intellect. Its main job will be to analyze the consequences of the ongoingderegulation of ocean shipping, which is both economically interesting and a matterof serious policy significance in its own right. The industry, more than a century oldand almost always federally regulated, is even now in the infancy of genuinecompetition for virtually the first time in its history, and study of its behavior offerscertain very interesting opportunities for inquiry.

However, because the original regulation of the industry and the near centuryof policy that ensued were driven almost exclusively by economic theory, a tacitquestion running throughout the study will be the merits of thinking about policywith theory or without it, or at least with something more than theoretical abstraction.Though perhaps a mere methodological issue of social science research, the conflictbetween theory and applied inquiry is an old and sometimes bitter one. Perhapssurprisingly, it is also quite politicized; historicism and case-specific empiricism havelong been associated with left-leaning political instincts.3 The problem is notimproved by this paper’s ultimate conclusion, which is that major policy decisions

4 Because the difference ultimately implicates very basic epistemological and metaphysicalquestions, it is as old ask the skeptical tradition itself. Cf. SEXTUS EMPIRICUS, OUTLINES OFPYRRHONISM (Julia Annas & Jonathan Barnes eds., 1994) (ca. 200 C.E.).

5 In American economic thinking it dates at least to the 1870s – and therefore it is at least asold as the synthesis of theoretical neo-classicism by Alfred Marshall. At the latest it began with therise of the German Historical School and its concern for full historical and institutional analysis ofparticular economic situations, and was imported directly to this country by a series of German-trainedeconomists who took university posts here in the 1880s. For a truly wonderful summary of thecompeting views, as well as the influence of German economic thinking on American universityeconomics, see DANIEL T. ROGERS, ATLANTIC CROSSINGS: SOCIAL POLITICS IN A PROGRESSIVE AGE76-111 (1998). See also HOVENKAMP, ENTERPRISE, supra note XXX, at 298-307.

6 Cf. Donald N. McClosky, The Rhetoric of Law and Economics, 86 MICH. L. REV. 752, 763(1988) (“But it must be evident that badness in talk is a result of being a bad person, like Hitler orTojo, not a result of using a bad method. Good or bad, we all use rhetoric. Law is rhetoric. Economicsis rhetoric. Mathematics is rhetoric.”).

7 A happy side effect of historicist inquiries is that their detailed study of seemingly dullsubject matter can prove to be more interesting than one could possibly expect. See, e.g., Allan W.Vestal, Public Choice, Public Interest, and the Soft-drink Interbrand Competition Act: Time to Derail

2

based on purely theoretical prognostication turned out to be quite questionable, andthat the history of shipping regulation illuminates the political or rhetorical use oftheory in policy debate.

But in any case, the conflict is an old one – depending on just how you lookat it, a really old one4 – and it is unlikely to be resolved anytime soon.5 So this paperfinds itself in a bit of a compromise. On the one hand, it would be wasteful to ignorethe fact that experience has been significantly at odds with the economic foundationsof traditional shipping policy, and that theory itself was a powerful and sometimesmisleading tool of political agitation. Interestingly, a careful historical look at thestate of the industry at the time shipping law was adopted, and the debate surroundingit since then, has something to say about how Congress works. But on the otherhand, this paper is probably not the place for any generalized theory of what is wrongwith theory. While the study here documents a history of self-interested abuse oftheory and the consequences of undue confidence in it, the paper is not really evena criticism of theorized policy as such. The apparently inevitable relegation ofeconomics to positivism, and a problematic positivism at that, is not theembarrassment that its critics and even some economists think it is.6 Accordingly,this paper’s approach, while richly historicist and context-specific, is driven largelyby current economic theory.7

the “Root Beer Express”?, 34 WM. & MARY L. REV. 337 (1993).I take heart in this fact, and hope you will do so as well.

8 ALFRED KAZIN, WRITING WAS EVERYTHING 13 (1995).

9 Indeed, why not “economics” as “literature”? See Gregory Scott Crespi, The Mid-LifeCrisis of the Law and Economics Movement: Confronting the Problems of Non-Falsifiability andNormative Bias, 67 NOTRE DAME L. REV. 231, 232 & n.6 (1991) (describing economics as a “literarygenre” in order “to emphasize my theme that law and economics writing is more appropriatelyregarded as a subjective, impressionistic, rhetorical, polemical form of literature than as a form ofscientific explication”); see also HOVENKAMP, ENTERPRISE, supra note XXX, at 7 (“There is nodichotomy between science and politics. On the contrary, science is a form of politics.”); McClosky,supra note XXX, at 763(“Economics is rhetoric. . . . [But] [r]hetoric is not a fault to be overcome.”).

An interesting analogy is in Richard Rorty, Philosophy as a Kind of Writing, in PRAGMATISM:A READER (Louis Menand, ed. 1997).

10 “Liner” shipping is an industry term of art which means regularly scheduled commoncarriage of cargo by sea, which is now by far the predominant means of ocean transport, but which hasonly existed since about the time of the Civil War. See AMOS HERMAN, SHIPPING CONFERENCES(1983).

11 See OECD, FINAL REPORT, supra note XXX, at 10; HERMAN, supra note XXX, at 3 (notingthat as of 1983 over 80% of world trade by volume was carried by ship).

12 Ninety-five percent of U.S. foreign commerce is transported in ocean-going vessels,roughly half of which is carried on vessels covered by the Shipping Act and which enjoy that Act’santitrust exemption. See Constantine G. Papavizas & Lawrence I. Kiern, 1997-98 U.S. MaritimeLegislative Developments, 30 J. MAR. L. & COM. 487, 488 (1999) (citing 144 CONG. REC. S11301(Oct. 1, 1998) (statement of Sen. Hutchison)).

3

Indeed, the most reassuring thing about economic theory is that its repeatedinadequacy leaves room for reasonable disagreement. Happily, uncertainty alsoensures a perpetually rich policy culture. Alfred Kazin said of literature that whatdrives the writer on is the radical insufficiency of language,8 and political normativityis no less kept living by its refusal to submit to quantitative mastery.9

* * *

Over its 140 year history, ocean liner shipping10 – by far the main mode ofinternational transport of goods11 and the chief means by which U.S. goods areshipped in foreign commerce12 – has almost always been directly regulated by theU.S. government. A chief feature of U.S. policy, which currently is embodied in the

13 Shipping Act of 1984, Pub. L. 98-237 (1984), 98 Stat. 67 (1984), codified at 46 U.S.C.App. §§1701-19, as amended by the Ocean Shipping Reform Act of 1998, Pub. L. 105-258, 112 Stat.1902 (1998).

14 See 46 U.S.C. App. §1701 (Shipping Act declaration of policy); See generally infra notesXXX and accompanying text.

15 The series of legislation commonly known as “deregulation” actually began in the middle1970s, with the Railroad Revitalization and Regulatory Reform Act of 1976, Pub. L. 94-210, Feb. 5,1976, 90 Stat. 31 (later strengthened by the Staggers Rail Act of 1980, Pub. L. No. 96-448, 94 Stat.1895 (1980)), the Airline Deregulation Act of 1978, Pub. L. 95-504, 92 Stat. 1705 (1978), and,somewhat less deliberately, the Public Utility Regulatory Policy Act of 1978, Pub. L. No. 95-617, 92Stat. 3117 (1978) (beginning deregulation of power industry). Deregulation of motor carriers quicklyfollowed. See Motor Carrier Act of 1980, Pub. L. 96-296, 94 Stat. 793 (1980). See generally DarrenBush & Carrie Mayne, In (Reluctant) Defense of Enron: Why Bad Regulation Is To Blame forCalifornia’s Power Woes (Or Why Antitrust Law Fails To Protect Against Market Power When theMarket Rules Encourage Its Use), 83 OR. L. REV. 207, 214-15 (2004) (discussing powerderegulation); Christopher Clott & Gary S. Wilson, Ocean Shipping Deregulation and MaritimePorts: Lessons Learned from Airline Deregulation, 26 TRANS. L. J. 205, 207-08 (1998) (discussingderegulation of transportation industries).

Though deregulation was begun during the Carter administration, it was a core mission of theReagan administration and much of the relevant effort was accomplished following Ronald Reagan’selection in 1980. See BARRY D. FRIEDMAN, REGULATION IN THE REAGAN-BUSH ERA: THE ERUPTIONOF PRESIDENTIAL INFLUENCE (1995); cf. Christopher C. DeMuth, Deregulation Review, 53 ANTITRUSTL. J. 189 (1984) (views of Reagan’s chief deregulatory officer within Office of Management andBudget).

16 Pub. L. 105-258, 112 Stat. 1902 (1998).

4

Shipping Act of 1984 (“Shipping Act”),13 has been exemption from antitrust forprice-fixing cartels of ocean carriers, premised on the belief that problems of cost andcapacity inherent in the trade can be resolved only by horizontal collusion. In placeof antitrust the industry has been regulated by a federal agency, now known as theFederal Maritime Commission (“FMC”). This policy’s purpose was to protect theindustry from perceived dysfunctions thought to inhere in its markets, and to preservea ready merchant marine for U.S. commerce and in times of national emergency.14

Always more or less controversial, this policy met the beginning of its endwith the adoption of the Shipping Act in 1984. At that time ocean shipping becameone part of the deregulatory effort central to the early Reagan administration.15 Amuch more significant change followed in the Ocean Shipping Reform Act of 1998(“OSRA”),16 and deregulatory efforts have not ended yet. For the first time in itshistory ocean shipping is rapidly becoming a free market.

17 That is, advocacy on behalf of the policy consisted of exclusively theoretical argumentsabout what would happen if the industry were ever subject to free competition.

18 See Stephen C. Pirrong, An Application of Core Theory to the Analysis of Ocean ShippingMarkets, 35 J. L. & ECON. 89 (1992); William Sjostrom, Antitrust Immunity for Shipping Conferences:An Empty Core Approach, ANTITRUST BULL., Summer 1993, at 419; see also William Sjostrom, LinerShipping: Modeling Competition and Collusion, in HANDBOOK OF MARITIME ECONOMICS ANDBUSINESS 307 (Costas T. Grammenos ed. 2002) [hereinafter “Sjostrom, Modeling Competition”].

19 See J.E. Davies, An Analysis of Cost and Supply Conditions in the Liner Shipping Industry,31 J. INDUS. ECON. 417 (1983); William Sjostrom, Collusion in Ocean Shipping: A Test of Monopolyand Empty Core Models, 97 J. POL. ECON. 1160 (1989); J.A. Zerby & R.M. Conlon, Joint Costs andIntra-Tariff Cross-Subsidies: The Case of Liner Shipping, 31J. INDUS. ECON. 383 (1983).

A body of positive empirical study also exists testing such matters as whether conferencemarket shares correlate with freight rates, compare N.R. Fox, An Empirical Analysis of Ocean LinerShipping, 19 INT’L J. TRANSP. ECON. 205 (1992) (finding such a correlation) with PAUL S. CLYDE &JAMES D. REITZES, FEDERAL TRADE COMMISSION, THE EFFECTIVENESS OF COLLUSION UNDERANTITRUST IMMUNITY 2 (1995) (finding no statistically significant correlation), whether rate variationis explained by particular cost factors, see Ingrid Bryan, Regression Analysis of Ocean Liner FreightRates on Some Canadian Export Routes, 8 J. TRANSP. ECON. & POL’Y 161 (1974); Fox, supra; TrevorD. Heaver, The Structure of Liner Conference Rates, 21 J. INDUS. ECON. 257 (1973); J. Janson & D.

5

The story of shipping deregulation, however, is so much more. For one thing,the pre and post-deregulation evidence now available presents one of the world’s rareopportunities for natural experiment on the behavior and effectiveness of collusivecartel pricing (rare because price-fixing is normally illegal and therefore done insecret). Likewise, because normal and effective competition really never existedprior to 1998, the normative foundation of the antitrust exemption was based almostentirely on theoretically modeled economic arguments.17 Observing the industry’sbehavior under deregulation is therefore a before-and-after opportunity to test thepredictive accuracy of at least one body economic argument, and among other thingsthis example happens to test an argument in industrial organization known as “emptycore” theory. Finally, shipping deregulation tells a larger historiographical story ofa deep evolution in U.S. competition policy.

This paper sets forth two major themes. First, deregulatory experience so farsuggests that liner markets can perform well under normal price competition,contrary to long-standing claims from the industry and some academics. The carriersthemselves have long argued that peculiar cost and capacity problems of their tradewould make it impossible for them to perform under competition, and prior toderegulation a body of theoretical work grew to support these claims and to arguethat they required antitrust immunity.18 Some empirical support was found as well.19

Shneerson, The Effective Protection Implicit in Liner Shipping Freight Rates, 50 REV. ECON. &STATISTICS 569 (1978), and whether conferences are able to price discriminate, see Fox, supra(finding evidence of discrimination); Bryan, supra (same).

20 See COMPTROLLER GENERAL, GENERAL ACCOUNTING OFFICE, REPORT TO THE CHAIRMAN,COMMITTEE ON MERCHANT MARINE AND FISHERIES OF THE U.S. HOUSE OF REPRESENTATIVES:CHANGES IN FEDERAL MARITIME REGULATION CAN INCREASE EFFICIENCY AND REDUCE COSTS IN THEOCEAN LINER SHIPPING INDUSTRY (1982) [hereinafter “GAO REPORT”]; George Deltas et al.,American Shipping Cartels int eh Pre-World War I Era, 19 RES. IN ECON. HIST. 1 (1999); J.W.Devanney III et al., Conference Ratemaking and the West Coast of South America, 9 J. TRANSP. ECON.& POL’Y 154 (1975); Heaver, supra note XXX.

21 See FEDERAL MARITIME COMMISSION, THE IMPACT OF THE OCEAN SHIPPING REFORM ACTOF 1998 (2001) [hereinafter “FMC, OSRA REPORT”]; DIRECTORATE OF SCIENCE, TECHNOLOGY ANDINDUSTRY, ORGANIZATION FOR ECONOMIC COOPERATION AND DEVELOPMENT, COMPETITION POLICYIN LINER SHIPPING: FINAL REPORT (2002) [hereinafter “OECD, FINAL REPORT]; see also CARSTONFINK ET AL., TRADE IN INTERNATIONAL MARITIMES SERV ICES: HOW MUCH DOES POLICY MATTER?World Bank Rep. No. WPS2522 (Jan. 31, 2001) (econometric study by economists withinDevelopment Research Group, World Bank). Another interesting government report is a very thoroughFederal Trade Commission empirical study prepared by two staff economists in 1995. While theyfound no statistically significant correlation between conference market share and freight rates, theydid find that freight rates were significantly lower where conference members were free to negotiateindividual contracts directly with shippers. See CLYDE & REITZES, supra note XXX, at 2.

Debate on these matters also contains less careful analyses, which often claim no more thanthat unrestrained competition is or is not desirable and do not invoke empirical support. See, e.g.,ADVISORY COMMISSION ON CONFERENCES IN OCEAN SHIPPING, REPORT 165-66 (1992) [hereinafter“ADVISORY COMMISSION REPORT”] (Separate Statement of Sen. Howard M. Metzenbaum) ; Statementof Charles James, Assistant Attorney General, U.S. Dept. of Justice, Antitrust Division, Before theHouse Committee on the Judiciary on H.R. 1253, The Free Market Antitrust Immunity Reform Actof 2001 (June 5, 2002); Statement of John Nannes, Deputy Assistant Attorney General, U.S. Dept. ofJustice, Antitrust Division, Before the House Committee on the Judiciary on H.R. 3138, The FreeMarket Antitrust Immunity Reform Act of 1999) (March 22, 2000).

22 See infra notes XXX and accompanying text.

6

Critique also predated deregulation, however,20 and grew after OSRA, notablyincluding studies by government and non-governmental organizations critical of theconference system in light of deregulatory experience.21 This recent work largelydrives the ongoing effort for complete deregulation, which seems likely to succeedin the near future.22

Second, the evolution of shipping policy is part of a larger history. On theone hand, it mirrors very closely the overall evolution of competition policy itself.So much is now different than it was in 1916, when the United States adopted its first

23 See, e.g., WORLD SHIPPING COUNCIL, A REVIEW OF “TRADE IN INTERNATIONAL MARITIMESERVICES” – A PAPER BY WORLD BANK RESEARCHERS (2001).

7

shipping policy – markets are different, industry is different, microeconomics is verydifferent, and both the law and the law’s esteem for the views of academiceconomists are different. As it happens the nature and prominence of statutorydisplacements of antitrust in favor of regulation is also very different, and as will beseen the story of the shipping exemption reflects the larger story of governmentefforts to cope with perceived of the problems of industrial organization.

What is particularly interesting about this history is what it says about the wayCongress works and about the use of abstract theory in the making of public policy.There exists a different and more plausible explanation of the plight of shipping thanthe one on which Congress based federal policy for nearly a century, and the reasonfor the continuing misunderstanding has been a predictive failure of economic theory.In short, a likely well-intentioned theoretical error by Congress, made at a time ofhistoric, world-wide economic distress, gave rise to a situation in which carriersmade substantial capital commitments to an inefficient market organization. Despitethe flaws in this situation, their interests predominantly lay in preserving that statusquo. As will be seen, it was quite easy for carriers and their academic supporters toexplain theoretically that this inefficient organization was actually a natural one, andto characterize the healthy process of forced exit that would result from deregulationas “destructive competition.” Those efforts persist even now, despite the weightyderegulatory evidence that the theory was inaccurate.23

The paper proceeds in six Parts. Parts I and II summarize the history of theocean shipping industry itself and the legal background of its regulation. Part II inparticular examines the American economic situation and the state of Americaneconomic thinking during the late nineteenth century, which, it will be argued,explain the real origins of the policy. Next, because shipping policy has always beendefended on the basis of the peculiar economic problems said to inhere in theindustry, Part III lays out its current economic situation, and such evidence as existsto support or contradict those claims. Parts IV and V then proceed to the heart of theinquiry, which is a critical examination of the a priori theoretical claims upon whichthe policy was traditionally defended. In particular, Part V addresses seemingly themost important question of theory, which is why, if the shipping cartels permitted byU.S. policy were in fact harmful or poorly behaved, they did not simply encouragedisciplinary entry or cheating by their own members. Finally, Part VI concludes withsome thoughts about whether this policy has lived up to its goals and theoreticalconceptions, and also about how the history of shipping regulation can be understood

24 The first American-built ocean-going sailing vessel was built in 1607, the same year as thefounding of the Jamestown colony. See Max Millikan, The Ocean Shipping Industry, in THESTRUCTURE OF AMERICAN INDUSTRY 422, 424 (Walter Adams ed. 1950).

25 In the days of sail, American shipbuilding had natural advantages. Abundant timber andother raw materials, good harbors, and a great demand for shipping services made shipbuilding aprincipal colonial industry, and by the early 18th century American built ships were the best andcheapest in the world. See Millikan, supra note XXX, at 424-25.

26 American ship operation also enjoyed special advantages. Prior to the Revolution,American shipping benefitted from Britain’s mercantilist economic policy – manufacturing wasrestricted to the home country, so there was heavy exchange of American raw materials across theAtlantic. Likewise, following the Revolution, American shipping benefitted greatly from theNapoleonic Wars, which kept the European belligerents occupied and left to American ships the creamof the neutral traffic. See MILLIKAN, supra note XXX, at 425; see generally Paul Stephen Dempsey,Transportation: A Legal History, 30 TRANS. L. J. 235, 269-70 (2003).

27 See Sjostrom, Modeling Competition, supra note XXX, at 2, citing GORDON H. BOYCE,INFORMATION, MEDIATION AND INSTITUTIONAL DEVELOPMENT: THE RISE OF LARGE-SCALEENTERPRISE IN BRITISH SHIPPING, 1870-1919, at 24 (1995) (noting that because sailing ships dependon wind, only steamships could provide regularly scheduled liner service).

28 First, the Civil War itself resulted in the destruction of a huge number of U.S. merchantvessels and also frightened owners into selling to foreign buyers, the combined result being a loss ofnearly half of the U.S. merchant fleet. Second, the War occurred at the same time that steam poweredocean shipping had become economically viable and had shown its superiority to sail in both speedand regularity of service. By the time the war ended and the economy in general began to recover,U.S. efforts to develop a domestic steamship industry were frustrated by having lost the historic

8

in the larger picture.

I. AN OVERVIEW OF OCEAN LINER SHIPPING

Virtually from the first European colonization American shipbuilding was asuccess,24 and until the Civil War American-built ocean vessels were preeminent.25

American ship operation was long an equivalent success.26 However, a collection ofevents conspired against both industries at about the time of the Civil War; bothdeclined dramatically at that time and neither has ever fully recovered. A chief factorin this collapse was the advent of the steamship and its ability, which had neverbefore existed, to provide regularly scheduled service.27 For a variety of reasons, atabout this time British shipbuilders and ship operators began far to outperform theirAmerican competitors,28 and in particular the natural advantages of American

advantage over inputs (which had been wood and highly skilled domestic workers) and by seriousbarriers to U.S. access to steel. The U.S. steel industry had not kept pace with its British counterpart,and the western expansion of the U.S. railroad network created a great demand for domestic iron. Thesame demands arising from the British railway and industrial revolutions had already stabilized.Finally, at the same time that these events had conspired to make American-built ships moreexpensive, capital was massively diverted by the attractive returns in the newly developing railroadand manufacturing endeavors of the American industrial revolution. See Millikan, supra note XXX,at 427-29; see also Dempsey, supra note XXX, at 270 (“By the dawn of the 20th Century, theAmerican merchant marine was nearly insignificant.”).

Even though domestic manufacturing later became more cost effective, by the time that hadoccurred U.S. ship operators faced a different problem – the rising cost of American labor. BecauseU.S. law had mandated the use of at least some American officers and crew on American-flag vesselssince 1789, see infra notes XXX and accompanying text, and because American operators necessarilyhad to employ more expensive American workers for shore-side support functions, U.S. shipping facedsignificant cost disadvantages from the turn of the 20th century that have never abated. See Millikan,supra note XXX, at 427-29.

29 See OECD, FINAL REPORT, supra note XXX at 18.


31 See HERMAN, supra note XXX, at 15; THOMAS J. SCHOENBAUM, ADMIRALTY ANDMARITIME LAW 489 (3d ed. 2001); GERALD H. ULLMAN, U.S. REGULATION OF OCEANTRANSPORTATION UNDER THE SHIPPING ACT OF 1984, at 4 (1995). The Shipping Act adopts a similar

9

building and operation of sailing ships were rendered moot. Thus it was that not longafter the Civil War the world fully entered the new era of “liner” shipping.

A point that will become important in the discussion to follow is that theearliest, completely unregulated and unorganized period of the middle nineteenthcentury is virtually the only time in its history that liner shipping has been subject togenuine competition. Though empirical evidence from that period is sometimesmarshaled in support of arguments about current policy, it should be remembered thatthe circumstances of that time were different than the present. First, the nascent linerindustry was in the midst of its own explosive initial growth, as new carriersstruggled for toeholds in the new market. Likewise, the industry faced competitionfrom the collapsing sailing ship industry, which could remain in business only bycharging desperately low rates.29 In any case, since the late nineteenth century bothAmerican ship building and ship operation depended on federal subsidization forsurvival.30

Likewise, virtually as long as there has been liner shipping there have beenconferences to govern it. A “conference” is any collection of carriers who by formalagreement decide to adhere to terms of service, including the fixing of rates.31 The

definition for the “conferences” to which it applies. See 46 U.S.C. App. §1702(7) (“ ‘conference’means an association of ocean common carriers permitted, pursuant to an approved or effectiveagreement, to engage in concerted activity and utilize a common tariff . . . .”).

32 See B.M. DEAKIN & T. SEWARD, SHIPPING CONFERENCES: A STUDY OF THEIR ORIGINS,DEVELOPMENT AND ECONOMIC PRACTICES 23 (1973); ULLMAN, supra note XXX, at 3-4. Theconference in question, the “England-Calcutta Conference,” was created at the urging of the steamshipleader Sir Samuel Cunard. See id. There is evidence of prototypical conferences existing earlier, asearly as 1850s, though they were not “modern” in the sense that they seem never to have agreed onprices or output. See Pirrong, supra note XXX, at 117; Sjostrom, Modeling Competition, supra noteXXX, at 3.

33 U.S. carriers followed the English example around the turn of the 20th century, seeULLMAN, supra note XXX, at 4, and the first major federal investigation of the phenomenon found thatby 1914 nearly every trade route in both the foreign and domestic commerce of the United Statespractically all established steamship lines had either been consolidated through stock ownership orwere working in cooperation through conference agreements. See Report of the Committee on theMerchant Marine and Fisheries on Steamship Agreements and Affiliations in the American Foreignand Domestic Trade, H.R. Doc. No. 805, 63d Cong., 2d Sess., at 415 (1914) (a document routinelyreferred to as the “Alexander Report” after its Chair, Rep. Joshua Alexander) [hereinafter“ALEXANDER REPORT”]. See also ROYAL COMMISSION ON SHIPPING RINGS, REPORT, Cmnd. No. 466819 (1909) [hereinafter “RING COMMISSION REPORT”] (major report of British investigative body,finding that as of the report’s publication in 1909 “the system of Conferences . . . ha[d] been extendedto many trades and over a wide area.”); see id. at 36 (“So far as rates, at least, are concerned,competitive services between the lines now in conference have ceased to exist: and, in general,competition from outside lines has been greatly restricted. The system has tended towards, andresulted in, a monopoly”); ADVISORY COMMISSION REPORT, supra, note XXX, at 5; Pirrong, supranote XXX, at 117 (“By the beginning of the twentieth century’s second decade, conferences existedon all major liner shipping markets.”).

34 See OECD, FINAL REPORT, supra note XXX, at 19.

35 Conferences once existed among passenger liners, and strictly speaking they are stillpermitted under U.S. law, but as a practical matter cheap and accessible air travel did away withpassenger liner service. See HERMAN, supra note XXX, at 77-78. Therefore, for practical purposes“shipping conference” means a conference of deep-sea general cargo liner carriers.

Shipping conferences exert little or no influence in coast-wise shipping, the bulk cargo trades,

10

first modern conference is believed to have been formed in 1875 over routes betweenEngland and India, at the behest of leading English carrier companies,32 andthereafter conferences rapidly spread to most of the main world trade routes.33 Todaythere are about 150 shipping conferences in the international freight trade,34 thoughactive conferences exist almost exclusively among liner services for deep-sea generalcargo transport.35 Conferences have dwindled in number both because of

or amongst tramp carriers. There had long existed conferences in the European coast-wise trade, butby the 1970s these were essentially defunct. See DEAKIN & SEWARD, supra note XXX, at 13. Cartelsin coast-wise shipping would not be permitted by U.S. law, which permits only conferences of shippersin “U.S. foreign commerce.”

Likewise, of the three general categories of ocean shipping (dry bulk, liquid bulk, andgeneral), conferences exist only among general cargo lines because most bulk shippers use either theirown vessels or third-party providers that are chartered exclusively to one shipper. Only general cargolines, by contrast, hold themselves out in the manner of common carriers, with regularly scheduledsailings available to all shippers at published tariff rates. “General cargo” includes all movements ofcargo except the major bulk commodities, which consist of oil, iron and manganese ores, coal, grain,bauxite, and phosphates. See generally ADVISORY COMMISSION REPORT, supra note XXX, at 17 &nn.1-3; HERMAN, supra note XXX, at 3; OECD, FINAL REPORT, supra note XXX, at 10, 19.

Finally, no conference of tramp carriers appears ever to have exerted influence. A “tramp,”as opposed to a liner, sails on an as-needed basis rather than according to regularly scheduled sailingsfor common carriage. A tramp or private carriage conference would not be permitted under U.S. law,which exempts only “common carriers” from antitrust. Tramp cartels have probably only existed inJapan, because they come within the shipping exception to Japanese competition law. See ADVISORYCOMMISSION REPORT, supra note XXX, at 36.

36 The effects of deregulation and consolidation are discussed infra at notes XXX andaccompanying text. The number of liner conferences has fallen by more than half in the past thirtyyears; as recently as 1973 there were about 360 conferences world wide. See DEAKIN & HOWARD,supra note XXX, at 13.

37 See OECD, FINAL REPORT, supra note XXX, at 22; Deltas et al., supra note XXX, at 67.

38 Container carriage is thus to be distinguished from traditional “break-bulk” carriage, whichwas shipment of cargoes that literally had to be broken down from the truck or rail transport thatbrought it to port, and then repacked in cargo ships. See ADVISORY COMMISSION REPORT, supra noteXXX, at 17 & nn.1-3.

11

deregulation and consolidation.36 Although there is a great diversity in the details ofindividual conference agreements, they are typically constructed around rate-fixingagreements. Some conference arrangements have gone so far as to allocate marketshares among conference carriers, and some have included rebates to shippers, jointservices, and door-to-door services.37

Liner shipping progressed under the conference system largely unchangeduntil the mid-twentieth century. Then, beginning in the 1950s and 1960s,transportation companies began devising means for the easy transfer of freight amongthe various traditional “modes” of transportation – that is, among truck, rail and ship.These efforts culminated in the development of “containerization,” in that it consistsof the use of large, standardized metal containers that can be carried either on specialsemi-truck trailers, flatbed rail cars, or specialized ocean vessels called“containerships,” and can be transferred easily amongst these “modes.”38 Cargo

39 See ADVISORY COMMISSION REPORT, supra note XXX, at 9 (noting that by 1960 labor costsin port accounted for 80% of the total cost of a typical voyage, and that following containerization theaverage handling time per voyage fell from 157 hours to 31 hours). Break-bulk shipping also createdlarge exposure to damage and pilfering. See id. at 17.

40 See ADVISORY COMMISSION REPORT, supra note XXX, at 17 & n.3; OECD, FINAL REPORT,supra note XXX, at 14. Note that ocean shipping cannot be fully containerized because there are somecargoes, such as liquid, bulk cargoes, and autos and other self-propelled equipment, that cannot beshipped by container. See ADVISORY COMMISSION REPORT, supra, at 17 n.3. These cargoes arenormally carried by charter or tramp, however, so it can be said that liner shipping is almost fullycontainerized.

41 Indeed, it is a basic maxim in transportation that unitization of cargo leads to increasedefficiency, speed of handling, and reduced costs. See Richard W. Palmer & Frank P. DeGiulio,Terminal Operations and Multimodal Carriage: History and Prognosis, 64 TUL. L. REV. 281, 284(1989).

42 See ADVISORY COMMISSION REPORT, supra note XXX, at 9; Palmer & DeGiulio, supranote XXX. Thus, the popular understanding of the organization of transportation services is currentlyat odds with its reality, and this result is largely to be blamed on conceptualization of “modes” oftransportation contained in traditional regulatory schemes. That is, transportation services arenormally conceived of as fitting in conceptually distinct “modal” cubbyholes such as “oceantransport,” “rail transport” or “truck transport,” and these distinctions are driven at least in part by thefact that the different “modes” have been conceived by regulators as different. For two extremelythoughtful and comprehensive historical analyses of this phenomenon, see Arthur Donovan,Intermodal Transportation in Historical Perspective, 27 Transp. L. J. 317 (2000), and Palmer &DeGiulio, supra.

12

handling prior to containerization was highly labor-intensive and costly39 and limitedoptimal ship size. Accordingly, it has come to pass that virtually all cargo in themajor U.S. trade lanes that can be moved by container does move that way.40

Containerization has led not only to cost savings and efficiency;41 its impact hasshifted to developing intermodal services – the efficient merging of differenttransportation modes into a seamless whole. Intermodalism has led to ever largerfirms providing through-transport and logistical services, which allow shippers toarrange for shipment from point of origin to point of destination by one economictransaction, on the basis of one set of international shipping documents, and alsoallows shippers to compare through-rates rather than the more complicatedcomparison amongst a set of possible unimodal legs of one shipment.42

Containership vessels themselves remain in continual evolution, mainly bygetting larger, and this too has had an impact on the organization of the industry. Themarginal cost of adding more containers to a containership already sailing is low, solarger containerships are generally more efficient than smaller ones. Accordingly


44 See OECD, FINAL REPORT, supra note XXX, at 17-18; Christopher Clott & Gary S.Wilson, Ocean Shipping Deregulation and Maritime Ports: Lessons Learned from AirlineDeregulation, 26 TRANS. L. J. 205, 213-15 (1998).

45 Already some ports, such as Singapore and Colombo, have developed specialized facilitiesto handle the “transshipment” of containerized cargo involved in such hub-and-spoke arrangements.Smaller ports, however, will be both unequipped to perform large scale transshipment operations andwill have inadequate harbors to handle the largest containerships. See id. Therefore, small ports mayincreasingly be reduced to regional feeder facilities for their nearest major ports. Likewise the growthin the size of containerships can only add to the rapid concentration of the industry, as the building ofcontainerships already entails enormous capital outlay. See infra notes XXX and accompanying text.

It has been suggested that the rise of hub-and-spoke arrangements may give rise to marketpower on behalf of carriers, similar to the market power thought to have inured to airlines through“hub” airport arrangements following airline deregulation. See Clott & Wilson, supra note XXX, at214. Whether this will occur remains to be seen.

46 A “service contract” is a contract between one or more shippers and one or more carriersor a conference in which the shipper commits to a certain volume of cargo over a fixed period of timeand the carrier commits to a certain rate and level of service. See 46 U.S.C. App. §1702(19);SCHOENBAUM, supra note XXX, at 404.

47 Prior to OSRA the Shipping Act had permitted service contracts, but severely restrictedtheir effectiveness. The Act: (1) permitted conferences to regulate or prohibit service contracting, (2)required that service contracts be filed with the Commission and made their terms available to thepublic, and (3) required that all “similarly situated” shippers be entitled to the same essential terms fora period of 30 days. See generally ADVISORY COMMISSION REPORT, supra note XXX, at 130-32.

Generally, prior to OSRA the conferences prohibited independent service contractingentirely. See FMC, OSRA REPORT, supra note , at 18; ADVISORY COMMISSION REPORT, supra note

13

shippers have continually added larger and larger ships.43 However, since very largecontainerships require huge capacity usage in order to operate efficiently, the use ofso-called “hub-and-spoke” arrangements, under which large ships serving main“trunk lines” are serviced by a number of smaller vessels operating regional feederroutes, are likely to continue their recent growth.44 This trend may pose significantconsequences for many port facilities, and will affect the organization of theindustry.45

Finally, what may prove as significant an event as containerization andintermodalism is a development that has only just begun, the incipient deregulationof ocean shipping. OSRA, which took effect in May of 1999, remains the mostsignificant step in U.S. policy thus far. Though it retained antitrust immunity forcarrier agreements, for the first time in the history of U.S. shipping policy OSRAmade it possible for any ocean carrier to negotiate independent “service contracts”with shippers,46 the terms of which may remain confidential, and which conferencesmay not deter in any way, either by action against any carrier47 or against any

XXX, at 133 (“At present, no conference in the U.S. foreign trades permits its member lines to engageindependently in service contract activity.”).

48 See 46 U.S.C. App. §1709(b)(3) (prohibiting retaliation against any shipper “because theshipper has patronized another carrier, or has filed a complaint, or for any other reason.”).

49 See FEDERAL MARITIME COMMISSION, 41ST ANNUAL REPORT FOR FISCAL YEAR 2002, at132 (2003) (48,154 new service contracts filed in fiscal 2002); FEDERAL MARITIME COMMISSION, 40THANNUAL REPORT FOR FISCAL YEAR 2001, at 124 (2002) (47,629 new service contracts filed in fiscal2001); FEDERAL MARITIME COMMISSION, 39TH ANNUAL REPORT FOR FISCAL YEAR 2000, at 112 (2001)(35,190 new service contracts filed in fiscal year 2000).

50 See ADVISORY COMMISSION REPORT, supra note XXX, at 1 33 & table VIII-1.

51 See FMC, OSRA REPORT, supra note XXX, at 20; see also OECD, FINAL REPORT, supranote XXX, at 22 (noting that since OSRA there has been “a rapid and massive switch (200%increase)” to service contracts, and that “[v]ery little traffic (e.g., less than 10% of the USA-Europetraffic) now takes place directly under conference terms.”).

Note that some portion of service contracts on file are actually between conferences andshippers or (conference service contracts differ from normal conference service because the contractsmay deviate from rates and terms in published tariffs and state rates that are fixed for a specified term),but the vast majority are independent – the FMC estimates that 98% of service contracts on file areindependent. See FMC, OSRA REPORT, supra note XXX, at 19.

52 See FMC, OSRA REPORT, supra note XXX, at 13-17. Demand will vary both accordingto (1) trade imbalances and currency fluctuation, and (2) underlying demand for particular cargoes.See infra notes XXX and accompanying text. The competitive environment of ocean shipping variessignificantly from place to place. The most competitive trade lane faced by shippers in U.S. foreigncommerce is between the east coast of the United States and Northern Europe. Under OSRA themembership of the conference that traditionally dominated that market, the Trans-Atlantic ConferenceAgreement (“TACA”), fell from a high of seventeen members to a current low of seven. See FMC,OSRA REPORT, supra note XXX, at 14-15. While TACA’s seven members still hold fifty percent ofthe market share of North Atlantic shipping (a drop from their pre-OSRA high of eighty percent), themarket is saturated by independent contract, and a huge portion of the transport even by TACA’smembers now moves under independent service contracts. See id.. U.S. shipping on the Pacific, bycontrast, remains comparatively uncompetitive. See id.

14

shipper.48 In the roughly seven years since OSRA’s enactment the service contracthas led to the virtual demise of the century-old conference system. The FMC nowreceives filing of nearly 50,000 new service contracts annually,49 a number that underprior regulation had been as low as 400 and tended to run between 2,000 and 4,000.50

The Commission estimates that as much as eighty percent of cargo carried byconference members is now carried under independent service contracts.51 Moreover,though OSRA still permits conferences to establish “voluntary guidelines” to governservice contracting, the evidence suggests that the ability of conferences to controlcarrier behavior through these guidelines is dependent on market conditions – wheredemand or competition are slack, the guidelines have comparatively little influence.52


54 See FMC, OSRA REPORT, supra note XXX, at 26.

55 It was once thought that in the case of liner shipping major concentration was unlikely fornon-economic reasons: concentration would result in the extinction of individual lines and may notbe acceptable to the governments of the flag states of the carriers involved. See Davies, supra noteXXX, at 433-34. Experience suggests this was incorrect.



58 See H.R. Rep. 98-53(II), 98th Cong., 1st Sess., 4 (1984).

59 See generally Clott & Wilson, supra note XXX (noting significant concentration in airlines,trucking and railroads following their deregulation in 1978 and 1980).

15

While it appears that the traditional conference system has collapsed underOSRA, that is not to say that carrier cooperation no longer has significance. Indeed,carriers continue to share price and customer information through so-called“discussion agreements,” and through information sharing and voluntary guidelinesthey likely have continued to exert some control over carrier behavior – particularlyin times of excess demand and particularly as to the various surcharges and ancillaryservice terms typical of contemporary service contracts.53 Moreover, carriers haveincreasingly turned to non-price operational agreements to organize their industry.In fact, operational agreements appear likely to replace the conference as theindustry’s key organizational tool; they now constitute the majority of agreements onfile with the FMC.54

Two salient consequences of recent deregulatory experience have been therapid concentration of the industry55 and the rapid transition to non-price operationalagreements among carriers and associations of carriers as a means of rationalizingcapacity. Since 1995 seven principal mergers and more than thirty acquisitions havetaken place.56 Thus, as of 2001 the top twenty liner operators accounted for 81% ofthe world fleet57 and 72% of world container capacity, and the five largest operatorsaccounted for 34% of capacity. Though this trend began even before the 1984 Act,58

it appears to have accelerated since OSRA. Very similar results followed thederegulation of other U.S. transportation sectors.59

Likewise, the transition to non-price operational arrangements indicates acoming reorganization of the industry. No longer able to rely on legally protectedconference ratesetting to stabilize capacity, carriers have turned to joint ventureagreements and various other arrangements for the sharing of vessels and leasing ofspace on one another’s vessels, as an alternative to simply adding new ships and


61 See S. Rep. 105-61, 105th Cong., 1st Sess., 2 (1997) (“All . . . maritime nations allowshipping conferences to exist with immunity from application of the antitrust or competition law.”).

Note that liner shipping in U.S. foreign commerce – the shipping to which the U.S. antitrustexemption applies – is not currently subject to any international regulation specific to shipping itself.A push for uniform international shipping regulation has long persisted, but it has been frustrated byparochial conflicts between developed countries and developing world. Many developing countries,upon emergence from colonialism following World War II, found themselves at severe balance-of-payments disadvantages and as net consumers of shipping services. See Terry Marquez, Shipping,Competition and Dumping: The European Community’s Liner Shipping Regulations, 23 TUL. MAR.L. J. 139, 150 (1998). Nearly 30 years ago, the United Nations Conference on Trade and Development(“UNCTAD”) attempted to address this conflict when it promulgated its Code of Conduct for LinerConferences, UN Doc. TD/Code/11/Rev. 1 (1974), reprinted at 13 I.L.M. 917 (1974). Thus, theUNCTAD Code includes not only rules for the conduct of conferences but also major “cargo sharing”provisions intended to support the development of merchant fleets by developing countries. Theseprovisions being unpalatable to maritime nations, the UNCTAD Code was ratified by many nations,but only with substantial reservations that have prevented it from being effectively implemented. SeeHERMAN, supra note XXX, at 164-74; Marquez, supra, at 150. Though support remains forinternational regulation, see, e.g., HERMAN, supra note XXX, at 173-74, this basic tension – theconflicting parochialisms of developed and developing countries – makes it seem unlikely that anytruly universal regulation will be forthcoming soon.

16

service strings, which tends to yet worse overcapacity problems. As the FMC wrotein a recent report, “[n]o longer can the structure of liner shipping be viewed as fiftyor so major carriers operating autonomously. It is more appropriate to view theindustry as blocs of operational partnerships, with crisscross ties via space chartersbetween many different members of different partnership blocs.”60

In summary, the rise of contemporary ocean shipping can be understood inhistorical perspective as a sequence of four watershed events: (1) the advent ofsteam-powered shipping in the 1870s, which made regularly scheduled shippingpossible for the first time and almost immediately gave rise to the conferences, (2)the advent of “containerization” in the 1960s, (3) a transition currently underway toan intermodal or amodal conceptualization of global transport services, with anattendant industrial and regulatory reorientation, and (4) a transition currentlyunderway to thorough deregulation, which has already resulted in largeorganizational changes in the industry.

II. LEGAL BACKGROUND

A. The Antitrust Exemption for U.S. Ocean ShippingFor almost as long as shipping conferences have existed, they have been

exempt from competition law both in the United States and elsewhere,61 and in most

62 See H.R. Rep. 98-53(I), 98th Cong., 1st Sess., 9 (1984) (“Generally, ocean shipping isunregulated in the rest of the world. For the most part, the U.S. is alone in having a regulatorycommission like the FMC”); GAO REPORT, supra note XXX, at iv (“Most countries do not imposerestrictions on the practices or organization structure of shipping conferences.”).

Thus, U.S. shipping policy is unique in at least three respects: First, it requires that allconferences be “open.” See 46 U.S.C. App. §1704(b)(2) (“Each conference agreement must . . .provide reasonable and equal terms and conditions for admission . . . to conference membership forany ocean common carrier willing to serve the particular trade or route.”). Second, U.S. law makesall conference agreements and tariffs approved thereunder available to the public. See 46 U.S.C. App.§1705(a) (“Within 7 days after an agreement is filed, the [FMC] shall transmit a notice of its filing tothe Federal Register for publication.”); id. at §1707(a)(2) (providing that conference “tariffs shall bemade available electronically to any person, without . . . limitation,” for a “reasonable charge”).Finally, since OSRA, U.S. policy has mandated that conference members be free to enter intoindependent contracts with shippers on a confidential basis. See 46 U.S.C. App. §1704(b)(8) (“Eachconference agreement must . . . provide that any member of the conference may take independentaction on any rate or service item upon not more than 5 calendar days’ notice to the conference”); id.at §1707(c)(2) (requiring “service contracts,” meaning “a written contract . . . between [a shipper] andan individual ocean common carrier . . .. in which the shipper . . . make s a commitment to providea certain volume of cargo over a fixed time period, and the ocean common carrier . . . commits to acertain rate [and] defined service level,” 46 U.S.C. App. §1702(19), “shall be filed confidentially withthe [FMC].”).

63 See U.S. v. Pacific and Arctic Ry. & Nav. Co., 228 U.S. 87 (1912); U.S. v. Prince LineLtd., 220 F. 230 (S.D.N.Y. 1915); U.S. v. Hamburg-Amerikanische Packet-Fahart-AktienGesellschaft, 200 F. 806 (S.D.N.Y. 1911). See generally HERMAN, supra note XXX, at 12. Indeed,the Justice Department began Sherman Act prosecutions almost immediately after the conferencesarose in the U.S. foreign commerce, even though they had already been found legal by the Britishgovernment, see ROYAL COMMISSION ON SHIPPING RINGS, REPORT, Cmnd. No. 4668 (1909)[hereinafter “RING COMMISSION REPORT”]; ULLMAN, supra note XXX, at 4, and by British courts, seeThe Mogul S.S. Co. v. McGregor, Gour & Co., 15 Q.B.D. 476 (1995); HERMAN, supra note XXX,at 8-9.

Even today, most conduct undertaken by conferences would be illegal were it not for theShipping Act exemption. Both common tariffs and the pricing “guidelines” typical of recent practicewould constitute per se illegal horizontal price-fixing. See WILLIAM C. HOLMES, ANTITRUST LAWHANDBOOK 285 (2004 ed., 2004). Allocation of customers or territories or restrictions on output (i.e.,“capacity rationalization”) likewise would be per se illegal. See id. at 322-23. Even mere exchangeof price and customer data, another carrier conference practice in recent years, could well be per seillegal under the circumstances in which the conferences participate. See U.S. v. Container Corp. ofAmerica, 393 U.S. 333 (1969); HOLMES, supra, at 292-93. Even were a conference re-conceived assome sort of joint sales agency, it would face significant antitrust risks. See U.S. Dept. of Justice andFederal Trade Commission, Antitrust Guidelines for Collaborations Among Competitors §3.2 (2000)(“labeling an arrangement a ‘joint venture’ will not protect what is merely a device to raise price orrestruct output; the nature of the conduct, not its designation is determinative.”).

17

of the world have gone completely unregulated.62 It so happens that liner conferencesarose in U.S. commerce at just about the time of the first U.S. antitrust legislation,and before Congress adopted the first U.S. shipping statute in 1916 the federal courtshad found their conduct illegal under the Sherman Act.63 However, conference

64 The British government appointed the Royal Commission on Shipping Rings, often calledthe “Ring Commission,” in 1906, to study the effects of liner conferences and their practices. TheRing Commission issued an influential report in 1909. See HERMAN, supra note XXX, at 11.

65 Some years after the Ring Commission issued its report, the House Merchant Marine andFisheries Committee undertook its own investigation of the conferences. The Committee, usuallyreferred to as the “Alexander Committee,” after its chair, Rep. Joshua Alexander, issued its report in1914. The report, still a basic instrument in shipping policy and the foundation for what wouldbecome the Shipping Act of 1916, is referred to as the Alexander Committee Report. See ALEXANDERREPORT, supra note XXX.

66 The Alexander Committee recommended the policy that became the 1916 Shipping Act. The Act followed the Committee’s report very closely, differing only in certain details. See EdwardMansfield, The Federal Maritime Commission, in THE POLITICS OF REGULATION 42, 46 (James Q.Wilson, ed. 1980). The Ring Commission, for its part, was more sanguine about the conferences andrecommended only that shippers’ organizations be formed to counterbalance the power of theconferences, and that the British Board of Trade keep conferences under review by requiring the filingof agreements and tariffs. See HERMAN, supra note XXX, at 11.

Ultimately, although both inquiries were concerned with suspicious ancillary conductdesigned to frustrate new entry or enforce conference terms, they sanctioned the existence of theconferences because they believed that without private coordination the industry would be subject tochaotic rate wars. In particular, both inquiries were interested in the use of “deferred rebates,” andother contract schemes to ensure shipper loyalty, and the use of so-called “fighting ships,” which wereships employed to forestall entry by non-conference carriers by underselling any would-be entrant,even at predatorily low prices. The fighting ship’s losses would be spread across a conference’smembers. See HERMAN, supra note XXX, at 11-13. By the time of the Alexander Committee Report,fighting ships had already been held illegal under the Sherman Act. See U.S. v. Hamburgh-AmericanS.S. Line, 216 F. 791 (S.D.N.Y. 1914).

The concerns of the government inquiries were predominantly economic, but were alsoseemingly moral as well. The Alexander Committee spoke of its desire to “eliminat[e] . . . secretarrangements and underhanded methods of discrimination.” H.R. Doc. No. 805, 63d Cong., 2d Sess.,416 (1914).

67 Pub. L. 87-254 (1916), 39 Stat. 728 (1916).

68 See generally ULLMAN, supra note XXX, at 4-6.

18

activity was the subject of two major government investigations near the turn of thetwentieth century, one in Britain64 and one in the United States,65 both of whichrecommended that conferences be tolerated but subject to some governmentoversight.66 Shortly following the U.S. investigation Congress passed the ShippingAct of 1916,67 which exempted liner conferences that operate in the U.S. foreigncommerce from federal antitrust for setting rates and other terms,68 though it requiredthat conference agreements be approved by the FMC’s predecessor before they were

69 For example, it prohibited price discrimination as between similarly situated shippercustomers, and prohibited artificial entry barriers such as the so-called “fighting ship.” See supra noteXXX.

70 Other than special provisions made during wartime, U.S. shipping policy has undergonethree major modifications since 1916. The first major amendment was made to undo a decision of theU.S. Supreme Court. In the Court’s decision in 1958 in Federal Maritime Board v. Isbrandtsen Co.,356 U.S. 481 (1958), the Court held “dual-rate” contracts illegal. This decision, which provedcontroversial, was reversed by Congress in 1961. See Pub. L. 87-346 (1961); see generally GAOREPORT, supra note XXX, at 7.

The second major modification came twenty years later in the Shipping Act of 1984.Onepurpose of the 1984 Act was to experiment with limited deregulation of the industry, as economistsand regulators had begun seriously to question the economic premises on which the industry’s antitrustexemption had been founded. See H.R. Rep. 98-53(II), 98th Cong., 1st Sess., 4 (1984). The Actbegan this deregulatory process by, among other things, introducing a limited form of servicecontracting.

The 1984 Act also addressed several concerns unrelated to deregulation, such as a measureto address the FMC’s long delays in approval of filed tariffs and carrier agreements, see infra noteXXX, and a provision to undo another controversial Supreme Court decision, FMC AktienbolagetSvenska, 390 U.S. 238, 243 (1968) (holding that the “contrary to the public interest” standard of the1916 Act required disapproval by FMC of any carrier agreement that would violate U.S. antitrust lawunless the carriers could carrey the burden of proving that the challenged agreement “was required bya serious transportation need, [and was] necessary to secure important public benefits or [was] infurtherance of a valid regulatory purpose fo the Shipping Act.”). See generally H.R. Rep. 98-53(II),98th Cong., 1st Sess., 3, 8-10 (1984); ADVISORY COMMISSION REPORT, supra note XXX, at 24-25;Palmer & DeGiulio, supra note XXX, 317-18.

Finally, the deregulatory innovation begun in the 1984 Act was significantly advanced whenOSRA became effective in May, 1999. See infra notes XXX and accompanying text.

71 Pub. L. 98-237 (1984), 98 Stat. 67 (1984), codified at 46 U.S.C. App. §§1701-19, asamended.

72 See 46 U.S.C. App. §1702(6) (defining “common carriers” to which Act applies as those“provid[ing] transportation by water . . . between the United States and a foreign country.”).

73 See 46 U.S.C. App. §1703(a) (defining “agreements” within the scope of the Act andtherefore exempt from antitrust upon filing with the FMC).

19

lawful, and prohibited certain practices thought to be harmful.69 Experience andchanging policy priorities have led to several modifications in this regulatoryscheme,70 culminating in the Shipping Act of 198471 which, as amended by OSRA,remains the law of the United States.

In short, current U.S. law permits liner carriers within the coverage of theAct72 to fix their rates and essentially any other terms of service or operations,73 so

74 See 46 U.S.C. App. §1704(a) (setting forth filing requirements). Agreements no longerneed be approved by the FMC. They become effective automatically 45 days after filing, see 46U.S.C. App. §1704 (providing that FMC can summarily reject agreements only for technical errors),a procedure modeled on merger pre-clearance under the Hart-Scott-Rodino Act, see H.R. Rep. 98-53(II), 98th Cong., 1st Sess., 10 (1984). This change grew out of frustration from all sides with delaysin the FMC agreement review process, which reached critical proportion in the 1970s and wasbelieved to have become an impediment to commerce. See id. at 7.

75 See 46 U.S.C. App. §1705(a) (requiring FMC to publish all filed agreements in FederalRegister). Though tariff rates no longer need be filed, conferences and carriers must still make thempublicly available in electronic form. See 46 U.S.C. App. §1707. A “tariff” is a schedule of rates forparticular services offered to the public on a common carriage basis. See SCHOENBAUM, supra noteXXX, at 493.

76 See 46 U.S.C. App. §1706.

77 First, the Act imposes a series of formal requirements for various kinds of agreements, suchas “reasonable and equal terms . . . for admission” to a conference and permission for conferencemembers to withdraw without penalty, see 46 U.S.C. App. §1704(b), and mandatory internalprocedural protections, see id. Second, the Act adopts a list of explicit prohibitions on anticompetitiveconference activities, such as unreasonable refusal to deal, predation against non-members, andallocation of customers, see 46 U.S.C. App. §1709(c).

78 Under §1705(g) the Commission can seek to enjoin any filed agreement if “the agreementis likely, by a reduction in transportation service or an unreasonable increase in transportation cost”to result in competitive injury. However, the Commission must seek such an injunction through anaction in the District Court for the District of Columbia; it cannot simply issue such an injunction onits own. See 46 U.S.C. App. §1705(h). The Commission can reject carrier agreements of its ownaccord only if they violate the technical requirements in §1704.

The FMC in fact has never used its power to challenge unreasonable service or costagreements as it exists under §1705(g) of the 1984 Act and OSRA, see ADVISORY COMMISSIONREPORT, supra note XXX, at 25, and §1705(g) may not be enforced by private action, see 46 U.S.C.App. §1710(a) (providing that private persons may file complaint with the FMC for any violation ofthe Shipping Act except for violations of §1705(g)). The Act also forbids third-party intervention inCommission actions under §1705(g), see 46 U.S.C. App. §1705(h), and prohibits private causes ofaction under the antitrust laws against any other conduct the Act prohibits, see 46 U.S.C. App.§1706(c)(2).

However, the Commission has worked with conferences on several occasions to modifytroublesome aspects of agreements, rather than proceed to litigation, and the Commission has longbelieved that overcapacity in U.S. foreign shipping lanes operates to constrain supracompetitive price

20

long as such agreements are first filed with the FMC.74 Upon filing the Commissionmakes them available to the public.75 Though filed agreements and activity relatedto them are exempt from antitrust,76 the Act adopts in its place something of asurrogate federal competition policy by way of imposing technical requirements onsome agreements and prohibiting certain conduct,77 and empowering the FMC toenforce a general minimum standard of competitiveness.78 There traditionally has

increases. See ADVISORY COMMISSION REPORT, supra note XXX, at 25.

79 See Mansfield, supra note XXX, at 46-61 (detailing evidence of FMC’s inability effectivelyto police filed agreements and tariffs according to any substantive standard).

80 See supra notes XXX and accompanying text.

81 The Shipping Act and OSRA each introduced other innovations unrelated to the antitrustexemption or contracting and rate-setting practices, notably provisions intended to encourage thetransition to multimodalism. See Palmer & DeGiulio, supra note XXX.

82 The Free Market Antitrust Immunity Reform Act (“FAIR Act”), first introduced as H.R.3138, 106th Cong., 1st Sess., in October of 1999 and re-introduced as H.R. 1253, 107th Cong., 1stSess., in March of 2001, would phase out the U.S. antitrust exemption as to all ocean shippingagreements except those among marine terminal operators, see H.R. 1253, 107th Cong., 1st Sess., §2;H.R. 3138, 106th Cong., 1st Sess., §2.

Indeed, even OSRA and the 1984 Act were compromise measures between carriers on theone hand and a variety of parties who sought to eliminate the antitrust exemption altogether. SeeDonald T. Bliss & David T. Beddow, Should the Shipping Act of 1984 Be Amended to EliminateConference Antitrust Immunity?, 36 FED. BAR NEWS & J. 357, 357-60 (1989); Papavizas & Kiern,supra note XXX, at 488-89.

83 The FAIR Act bill has had the support of two House Judiciary Chairmen and the AntitrustDivision under the administrations of both Presidents Clinton and George W. Bush. The bill wasintroduced in the 106th Congress by then-Chairman Hyde and in the 107th by current ChairmanSensenbrenner. As for the current views of the Antitrust Division, see Statement of Charles James,

21

been doubt, however, about the agency’s ability to enforce these rules. It seemslikely that throughout most of its history the industry has been subject to very littlepractical constraint on anticompetitive activity.79

Though in its structure and superficial details this regime resembles prior law,Congress has dramatically redesigned shipping policy in the past several years,mainly by way of OSRA. OSRA’s chief innovation was the introduction ofindependent service contracting, as to which conferences are not permitted to takeany punitive action, and the terms of which may be kept confidential.80 Thus, thoughin principle conferences may still collectively fix rates and other terms, theiragreements are no longer binding. In other words, as a practical matter, U.S.shipping is now a free market, except for the fact that carriers are not subject toantitrust. Accordingly they may share price information, agree to non-binding“guidelines” for rates and terms of service, adopt common non-binding tariffs, andso on.81 It seems unlikely that deregulation will end with OSRA. An effort has beenmade several times since 1999 to do away with the U.S. antitrust exemption almostentirely,82 and though it has never yet reached floor consideration, it has hadsignificant support.83 Certain other influential calls for end to the exemption have

Assistant Attorney General, U.S. Dept. of Justice, Antitrust Division, Before the House Committeeon the Judiciary on H.R. 1253, The Free Market Antitrust Immunity Reform Act of 2001 (June 5,2002) (supporting FAIR Act); Statement of John Nannes, Deputy Assistant Attorney General, U.S.Dept. of Justice, Antitrust Division, Before the House Committee on the Judiciary on H.R. 3138, TheFree Market Antitrust Immunity Reform Act of 1999) (March 22, 2000) (same).

As it happens, the Administration’s opinion concerning the Shipping Act exemption hasvaried over time. Compare U.S. Dept. of Justice, The Regulated Ocean Shipping Industry (1977)(critical of the exemption) with GAO REPORT, supra note XXX, at App. I (reprinting, as an Appendix,a Justice Department letter urging expansion of the Shipping Act exemption to permit “greaterfreedom to engage in restrictive business practices.”), but appears now to be strongly in favor of totalderegulation.

84 See OECD, FINAL REPORT supra note XXX; FINK ET AL., supra note XXX.


86 See Dempsey, supra note XXX, at 270-72.

87 See supra note XXX.

88See Edward Mansfield, The Federal Maritime Commission, in THE POLITICS OFREGULATION 42, 47-48 (James Q. Wilson, ed. 1980). As Mansfield notes, the war led to drasticincreases in shipping rates to freight in U.S. trades, The bill was approved along almost exclusively party lines, being voted for almost exclusively byDemocrats and opposed almost exclusively by Republicans. See id.

22

also emerged, most notably from the OECD and from economists at the WorldBank,84 and even now several other maritime nations are reconsidering their ownliner shipping exemptions.85

B. The Historical Setting and the Legislative Purpose Testing the sufficiency of this regime, among other things, calls for a bit

closer examination of what its purpose might be. As mentioned, of a necessity U.S.shipping policy was almost always based on theoretical speculation as to what wouldoccur if the industry were subject to open competition. It had to be theoreticalbecause the industry hardly ever faced such a thing.

One might think the most important fact surrounding the history of shippinglegislation is that it was first adopted during World War I. Shipping markets wereseverely disrupted by the war, and other U.S. transportation sectors had so severelysuffered that they were nationalized during the hostilities.86 The war, however, seemsto have played little role in debate on the substance of shipping policy. The reportof the so-called Alexander Committee87 predated the onset of hostilities, and so didnot mention it, and the 1916 Shipping Act followed the Committee’srecommendations very closely. A good case has been made that the crisis of the warled to the political alignments in Congress which made passage of the Act possible,88

89 See generally Tony Freyer, The Sherman Antitrust Act, Comparative Business Structure,and the Rule of Reason: American and Great Britain, 1880-1920, [CITE] at 994 (1989); JamesLivingston, The Social Analysis of Economic History and Theory: Conjectures on Late NineteenthCentury American Development, 92 AM. HIST. REV. 69, 72-73 (1987); Jeffrey G. Williamson,Watersheds and Turning Points: Conjectures on the Long-Term Impacts of Civil War Financing, 34J. ECON. HIST. 636 (1974) [hereinafter “Williamson, Watersheds”]; Jeffrey G. Williamson, LateNineteenth-Century American Retardation: A Neoclassical Analysis, 33 J. ECON. HIST. 581 (1973)[hereinafter “Williamson, Late Nineteenth-Century”].

90 See Williamson, Late Nineteenth-Century, supra note XXX.

91 See generally ALFRED D. CHANDLER, JR., THE VISIBLE HAND: THE MANAGERIALREVOLUTION IN AMERICAN BUSINESS 122-44 (1977); Robert L. Rabin, Federal Regulation inHistorical Perspective, 38 STAN. L. REV. 1189, 1199-1200 (1986).

92 See generally NAOMI R. LAMOREAUX, THE GREAT MERGER MOVEMENT IN AMERICANBUSINESS, 1895-1904 (1985).

23

but the war itself and its disruptions to shipping did not affect the thinking of theCommittee. The real intellectual roots of shipping policy appear to lie further back,in the economics and history of nineteenth century business.

First, times were economically bad and had been for some time. Economiesthroughout the entire North Atlantic were under severe stress during the last quarterof the century. A series of depressions had wracked the United States since the CivilWar, and industry struggled with a long list of economic problems.89 Of particularrelevance was concern over the power of organized labor. In the view of industry,whether correct or not, that influence – along with labor’s simultaneously rising costand decreasing productivity – was dramatically on the increase.90 Thus it becamewidely believed among industrialists that the only variable still in their control wasoutput, and the only way to regulate output, especially in industries carrying highfixed charges, was through industry cooperation.91

As a separate theme, debate on business policy following the Civil War waslaced with concern for the newness of circumstances. While this rhetoric of the newwas surely misleading in some respects – just as it has been in our own times – interms of the organization of business probably no other period in U.S. historywitnessed quite so many major changes in such a short time. It was a time of someof the most dramatic demographic changes in history, a fact that contributed to laborproblems, and it was also a time of rapid technological advance, rapid increases inscale of production, and rapid westward expansion. One of the most dramaticchanges of the period was the so-called “Great Merger Wave.”92 During the decadebetween 1895 and 1904, more than 1800 firms disappeared into consolidations. Theresulting combinations in many cases came to hold large market shares in their

93 See LAMOREAUX, supra note XXX, at 2-5.

94 See ALEXIS DE TOCQUEVILLE, DEMOCRACY IN AMERICA 513-17 (George Lawrence, trans.Doubleday Anchor Books ed. 1969) (1839) (“Americans of all ages, all stations in life, and all typesof disposition are forever forming associations. . . . In every case, at the head of any new undertaking,where in France you would find the government or in England some territorial magnate, in the UnitedStates you are sure to find an association.”).

95 J. WILLARD HURST, LAW AND SOCIAL PROCESS IN UNITED STATES HISTORY (1960).

96 See generally CHANDLER, supra note XXX, at 122-44; Robert L. Rabin, FederalRegulation in Historical Perspective, 38 STAN. L. REV. 1189, 1199-1200 (1986).

97 See 1 EARL W. KINTNER, THE LEGISLATIVE HISTORY OF THE FEDERAL ANTITRUST LAWSAND RELATED STATUTES 7 (1978); WILLIAM LETWIN, LAW AND ECONOMIC POLICY IN AMERICA: THEEVOLUTION OF THE SHERMAN ANTITRUST ACT 10 (1965).

98 See ROBERT F. HIMMELBERG, THE ORIGINS OF THE NATIONAL RECOVERY ADMINISTRATION:BUSINESS, GOVERNMENT, AND THE TRADE ASSOCIATION ISSUE, 1921-1933 (1993); Ellis W. Hawley,Herbert Hoover, the Commerce Secretariat, and the Vision of an “Associative State,” 1921-1928, 61J. AM. HIST. 116 (1974).

99 Indeed, business associationalism was a major article of faith with many business peopleat least as early as the legislative consideration of the 1914 antitrust amendments. Famously, the“Gary dinners,” informal gatherings of steel industry executives to coordinate business affairs arrangedby Elbert Gary, had begun in 1907 and were well known to the public at that time. See JOHN D.CLARK, THE FEDERAL TRUST POLICY (1931). Likewise, a theoretical work by the antitrust lawyerArthur Eddy, which was nothing short of a blockbuster and lay the intellectual foundations for theassociationalist movement that was to culminate in the National Industrial Recovery Act, waspublished and widely read in 1912. See ARTHUR JEROME EDDY, THE NEW COMPETITION (1912); seegenerally H.R. Tosdal, Open Price Associations, 7 AM. ECON. REV. 331 (1917) (describing Eddy’sbook and the impact it had on businesspeople’s thinking).

In any event, testimony surrounding the 1914 antitrust amendments involved a fair bit of this

24

industries, and while many of these shares were transitory, some of them persisted.93

Finally, a theme running behind all these circumstances is a murkiersociological phenomenon. That theme was what might be called “associationalism.”As de Tocqueville famously observed, Americans have a predilection to addressproblems through private associations of effort94; Willard Hurst, with less optimism,said that this instinct often manifested itself as a kind of “bastard pragmatism.”95

Among business people this associationalism had grown to dominance during the lasthalf of the nineteenth century,96 and indeed the felt tension between competition andcooperation explains much of the political tension behind the adoption of antitrustlaw.97 Associationalism would reach its zenith later, during the Presidency ofHerbert Hoover and the subsequent National Industrial Recovery Act,98 but by 1916it had become a prominent feature of policy debate99 and had already resulted in the

thinking, and included explicit requests that any modifications to the Sherman Act permit either openfreedom to fix prices or at least provide for a regulatory agency that could pre-approve some formsof cartel agreement. See CLARK, supra at 139-64.

100 See Tosdal, supra note XXX, at 331-34.

101 See generally ALFRED D. CHANDLER, JR., THE VISIBLE HAND: THE MANAGERIALREVOLUTION IN AMERICAN BUSINESS 122-44 (1977); HOVENKAMP, supra note XXX, at 145-48;Rabin, supra note XXX, at 1199-1200.

102 See Rabin, supra note XXX, at 1214.

103 See Rabin, supra note XXX, at 1200 (“[T]he railroads sought rationalization of the marketthrough pooling arrangements. They undoubtedly would have preferred an essentially passiveregulatory system that simply approved private pooling. Yet the railroads’ self-interest almostcertainly would have put them behind active governmental intervention that sanctioned cooperativeactivity”).

104 ALEXANDER REPORT, supra note XXX, at 418 (“Nearly all of the steamship linerepresentatives expressed themselves as not opposed to government supervision.”). Admittedly, thereis contemporaneous evidence of carriers’ opposition to the Shipping Act. See Mansfield, supra noteXXX, at 46-47. However, their opposition appears to have had less to do with the federally overseencartel structure than with the fact that the original shipping legislation called for the FMC’spredecessor to build, own and operate a fleet of commercial ships of its own. Not only were thecarriers opposed to new competition, they feared that simultaneous ownership and regulation wouldcreate a conflict of interest in their regulator. See id.

Indeed, the carriers could hardly fail to have supported U.S. oversight of some kind, sinceunregulated conference activities had already been found illegal under the Sherman Act. See supranote XXX.

25

spread of significant numbers of purportedly legal trade associations for thecoordination of business.100

These background facts – persistent economic malaise, the overall newnessof events, by far the largest consolidation of business in world history, and thegeneral atmosphere of business cooperation for the greater good – would surely havebeen in the air during the creation of the first U.S. shipping policy. It seemsilluminating that some time before the first shipping cartels were formed, Americanrailroad entrepreneurs had engaged in substantial efforts at collective self-regulationof their industry, citing the same concerns as the ship owners alter did.101 Likewise,even in early practice under the Interstate Commerce Commission the railroadscreated “rating bureaus” that, among other things, established pooling and capacityrationalization arrangements.102 The railroads very likely would have supportedgovernment sanction of their efforts even if it meant some constraints on theirbehavior.103 The view among ocean liner operators was the same.104

So, against this backdrop, with what theoretical apparatus did Congress

105 See ALFRED MARSHALL, PRINCIPLES OF ECONOMICS (1890).

106 See [CITE]

107 See HANS THORELLI, THE FEDERAL ANTITRUST POLICY: ORIGINATION OF AN AMERICANTRADITION 120 (1955) (“In fact, [the economists’] impact on public attitudes towards the trustproblem at the time the Sherman Act was passed appears to have been almost imperceptible.”)

108 During consideration of the 1914 amendments, congressional committees for the first timebegan taking the views of professional economists, as well as economically inclined lawyers andpolicy entrepreneurs, like Louis Brandeis. However, a predominant source of input remainedtestimony as to the practical experience of businessmen and men of affairs. See CLARK, supra noteXXX, at 139-64.

109 See ALEXANDER REPORT, supra note XXX, at 2-7 (listing the Committee’s sources ofinformation).

110 Interestingly the British were more charitable toward academic economists at this time,and the Ring Report discusses economic thinking with respect and concern. See, e.g., RINGCOMMISSION REPORT, supra note XXX, at 38. What is especially interesting is that, unlike the U.S.Congress, this academically informed British government upon its first consideration recommendednot only toleration of the conferences, but no regulation of ocean shipping whatever. See id. at 81-90(recommending formation of shippers’ councils and ongoing information collection, but rejecting anyactual regulation).

26

approach the shipping problem? The 1916 legislation was unquestionably driven by“economic theory,” but knowing just what that means requires some care. First,while Alfred Marshall had synthesized neoclassical price theory more than twodecades earlier in his influential Principles of Economics,105 and while his newmarginalism had champions in America,106 it appears that Congress simply didn’tcare. Throughout its legislative efforts concerning major aspects of Americancommerce from 1890 through the late Progressive era, Congress demonstrated opendisregard for economic science, at least as practiced in universities and academicjournals. Congress’s lack of concern is especially well documented as to theadoption of the Sherman Act in1890107 and, to a lesser degree, in the subsequentantitrust laws in 1914.108 For its own part, the Alexander Committee sought inputfrom no economists at all, but rather took in volumes of written and testimonialinformation provided by carriers and shippers.109 Nowhere in the Committee’s 459-page report does it cite a work of academic economics.110

Thus, while members of Congress and the various constituencies surroundingthe original shipping legislation plainly possessed their own economic theory, itlikely wasn’t neoclassical marginalism. Rather, the economics within Congress wasthe ad hoc, instinctive economics of working policymakers of the day, and itconsisted predominantly in lingering nineteenth century Jacksonianism. That is to

111 See generally HOVENKAMP, ENTERPRISE, supra note XXX; Donald Dewey, The EconomicTheory of Antitrust: Science or Religion?, 50 VA. L. REV. 413, 415 (1964); James May, Antitrust inthe Formative Era: Political and Economic Theory in Constitutional and Antitrust Analysis, 1880-1918, 50 OHIO ST. L. J. 244, 269-81 (1989).

112 See F. M. Scherer, Efficiency, Fairness, and the Early Contributions of Economists to theAntitrust Debate, 29 WASHBURN L. J. 243, 254-55 (1989) (“neither Congress nor the main body ofprofessional economists in the United States placed much discernible emphasis on allocative efficiency. . . . Deeply-rooted standards of fairness . . . had much to do” with economic policymaking in theday).

113 See ALEXANDER REPORT, supra note XXX, at 24-25; infra notes XXX.

114 See, e.g., ALEXANDER REPORT, supra note XXX, at 300 (noting concern expressed intestimony for rate discrimination as between shippers).

115 See, e.g., ALEXANDER REPORT, supra note XXX, at 21-22 (“[W]here[ver] several traderoutes adjoin or intersect one another the several conferences or groups of lines have entered intoagreements with each other, so that practically the entire traffic is being operated by the great majorityof the lines on the basis of established rules and in the spirit of harmony.”).

116 See ALEXANDER REPORT, supra note XXX, at 416 (“To terminate existing agreementswould necessarily bring about one of tow results: the lines would either engage in rate wars whichwould mean the elimination of the weak and the survival of the strong, or, to avoid a costly struggle,they would consolidate through common ownership.”).

The Committee apparently did not consider whether the behavior of such a monopolist couldbe constrained by potential competition.

27

say, it drew not only on classical political economy, but also on an explicitly moral,libertarian political philosophy, and the two were interwoven to a degree that theycannot be meaningfully disaggregated.111 Even academic economists in America haddeveloped no theoretical conception of allocative efficiency until a few decades intothe twentieth century,112 and accordingly neither had popular economic thinking.Thus, the Alexander Committee never mentioned any explicit efficiency goal, andinstead expressed its concern for “fairness,” both as to rates and methods ofcompetition,113 and as to equality of treatment for different shippers.114 TheCommittee likewise placed great emphasis on “harmony” within shipping markets,and by that the Committee plainly did not mean the absence of violence or laborunrest or the like. Rather, it meant “fair” competition.115

The Committee’s chief stated concern was concentration. The Committeebelieved that in the absence of voluntary cooperative agreements, competition wouldlead to monopolization.116 The Committee’s only stated basis for this view was itsreview of the evidence it had received; it cited no theoretical work in this connectionand did not explain in any theoretical manner why it believed monopolization wouldarise. But the mixed theoretical lineage of the Committee’s economics was on full

117 See, e.g., ALEXANDER REPORT, supra note XXX, at 24-25 (“The consensus of thetestimony before the Committee has been to the effect that the only practicable method of securinga fair share of business for all the lines, thus preserving the weaker ones, and of substituting harmonyin place of discord and cutthroat competition, is through some form of pooling as just described, andthat the primary purpose of such pooling agreements is to maintain a reasonable price and prevent theline or lines with unlimited capital for development from ultimately acquiring the whole business.”)(emphasis added).

118 See ALEXANDER REPORT, supra note XXX, at 295 (“Practically all steamshiprepresentatives who testified before the Committee, as well as a majority of the leading Americanexporting and importing firms . . . contended that shipping agreements . . . are a natural evolution andare necessary if shippers are at all times to enjoy ample tonnage and efficient, frequent, and regularservice at reasonable rates.”).

119 See RICHARD T. ELY, OUTLINES OF ECONOMICS 199 (1916) (discussing ruinouscompetition in railroads); Eliot Jones, Is Competition in Industry Ruinous?, 34 Q. J. ECON. 473(1920); see generally LAMOREAUX, supra note XXX, at 50 & n.3 (citing academic literature from asearly as 1918); cf. Washington Notes, 24 J. POL. ECON. 1012, 1013 (1916) (noting passage of 1916Shipping Act and noting the “wide difference of expert opinion” as to whether “transportation by wateris necessarily controlled by considerations very different from those that prevail in connection withrailway transportation”).

120 See LAMOREAUX, supra note XXX, at 50.

121 See generally Dempsey, supra note XXX, at 281-89. The trucking industry in particularwas thought, on similar economic reasoning, to be at risk of catastrophic overcapacity if exposed topure competition. See id. at 281-82.

28

display. On the one hand, the Committee thought that the inevitable concentrationwould arise through devious or violent trade practices, but on the other hand thenormative basis for avoiding that result was that some shipowners would be deprivedof their “fair” share of the business.117

Having said all that, however, it seems likely that the Committee receivedsome dose of prevailing orthodoxy in the testimony it received from industry. Manyshippers and, with uniformity, the carriers themselves, believed that collusion wasa natural condition of liner economics and was necessary in order to preserve theindustry’s health.118 For what it is worth, arguments from high fixed costs weretheoretically well developed by 1916, and were given sophisticated theoreticaltreatment in the academic literature.119 Moreover, there is evidence that industrialistsshared the view.120

Interestingly, the early twentieth century saw very similar patterns of publicpolicy in other transportation sectors, and accordingly similar regulation early arosein trucking and aviation.121 By the late 1920s and 1930s, a time when the federalgovernment had become more sympathetic to the views of academic economists, itwas largely economic orthodoxy that transportation industries could not function

122 See id. at 283.

123 In fact, ocean shipping is among the last of the industries traditionally regulated in thedeveloped world to face deregulation, see OECD, FINAL REPORT, supra note XXX, at 54, and theindustry appears to be following a now familiar pattern, see id. at 68 (“Many capital intensiveindustries (e.g., the power-generating sector, telecommunications, rail and air cargo transport) havein the past argued that they should retain the sam anti-trust immunity for price-fixing still afforded tothe liner shipping sector. . . . In many of these, suppliers sought to band together and set pricesamongst themselves for the same supposed ‘greater benefit’ of stability over economic efficiency.[And yet] in virtually all of these sectors, governments have seen fit to prohibit price-fixing amongcompetitors. Firms have come up with other, more market-oriented strategies to resolvesupply/demand imbalances and consumers have benefitted from greater and truer competition.”).

124 The U.S. government began subsidizing domestic shipbuilding and operation almostimmediately after independence and it has continued ever since. By a law of 1789 American registrywas restricted to vessels owned by U.S. persons. The Navigation Act of 1817 reserved the coastaltrades for American flag vessels, and mail subsidies were established in 1845. At the time of theirenactment, none of these protections or subsidies was really needed for protectionist purposes, becauseof the continued native advantages of the American industry, but rather were retaliation for similarpolicies of other maritime nations. In fact, they were long bitterly opposed by American shipoperators, who wanted access to the foreign-built ships that had become cheaper due to the rising costof U.S. labor. At about the time of World War I, when the shipping industry’s economiccircumstances had changed significantly, ship operating interests finally gave up their effort for sucha “free ship” policy and turned instead to seeking direct federal subsidization of ship operation. See

29

under competition.122

In any event, the theoretical thrust of shipping policy throughout its historyhas been, in substance, essentially similar. While theoretical economics, since 1916,has added sophistication to the “destructive competition” thinking of the earlytwentieth century, its gravamen has remained largely the same. The content of theseviews will be detailed presently in Parts III and IV.

III. CURRENT ECONOMIC CIRCUMSTANCES OF OCEAN LINER SHIPPING

As has been the case in other regulated industries, particularly at times ofproposed deregulation, the liner carriers have defended their special legal status onthe basis of economic dysfunctions allegedly inherent in their trade.123 The sectionsthat follow will discuss the various dysfunctions, including a set of purely artificialproblems not much discussed by the industry – the major government subsidies andother regulatory distortions that support it, which are widely employed by maritimenations for political rather than economic reasons.

A. Subsidization and Other Regulatory Distortions. Both shipbuilding and ship operation have long been subsidized by the U.S.

government, as they are in many countries,124 for domestic protectionist reasons, for

generally Millikan, supra note XXX, at 426-31.

125 See ADVISORY COMMISSION REPORT, supra note XXX, at 68-69.

126 To be a “U.S. flag vessel” means simply that a vessel is “documented” with the U.S.government. U.S. flag status entitles a ship owner to benefits under U.S. law as well as restrictionson ownership, operation and transfer. See generally Constantine G. Papavizas, New Developmentsin U.S. Flag Vessel Financing and Citizenship Requirements, 24 TUL. MAR. L. J. 205 (1999).Documentation is governed by the Vessel Documentation Law, 46 U.S.C. App. §§12101-24.

127 U.S. protections for U.S. maritime labor have also made U.S. flag operation so costly thatmany vessels that would otherwise qualify for U.S. documentation to register elsewhere, a resultseemingly at odds with the core policies of maritime protectionism. This point should be taken withcaution, however, as the fleet of ships under the control of U.S. persons is larger than the “U.S. flagfleet.” Though the U.S. flag fleet has declined significantly as a portion of worldwide capacity –between 1970 and 1997 the U.S. flag merchant fleet declined from fifth to eleventh largest worldwide,see Timothy Semenoro, Comment, The State of Our Seafaring Nation: What Course Has CongressLaid for the U.S. Maritime Industry?, 25 TUL. MAR. L.J. 355, 366 (2000), and as of 1997 only 3.9%of the international trade of the United States by weight was carried in U.S. flag vessels, see id. at 366– that is largely because U.S. controlled vessels have increasingly sought documentation in morepermissive countries, notably Liberia and Panama. [CITE]


129 See 46 U.S.C. App. §1177.

130 See 46 U.S.C. App. §§1271-80a; see generally SCHOENBAUM, supra note XXX, at 501.

131 See 46 U.S.C. App. §§1187-87e; SCHOENBAUM, supra note XXX, at 500-01. Currently47 vessels participate in this program. See id. at 501.

30

the felt need to preserve ready access of domestic industry to merchant marine, andfor national security reasons.125 However necessary they may be to overall policyobjectives, U.S. subsidy and protectionist policies have made U.S.-flag linershipping126 more costly than it might otherwise be,127 and have probably contributedto overcapacity.128

The federal government subsidizes new building at U.S. shipyards and newpurchases by U.S. citizens through tax deferment on capital construction funds129 andby guaranteeing financing of new construction in U.S. yards.130 The governmentprovides direct operating subsidies for certain U.S. flag vessel owners if they agreeto make their vessels and infrastructure available in time of national emergency.131

The government also subsidizes U.S. flag operators by restricting all coastwise cargo

132 This is provided for under the so-called “Jones Act,” section 27 of the Merchant MarineAct of 1936, 46 U.S.C. App. §883. “Coastwise” trade is carriage between U.S. ports or between aU.S. port and a U.S. flag vessel. See id. Strictly speaking, the Jones Act imposes stricter citizenshiprequirements than other U.S. flag policies. “Jones Act citizenship” requires both that the ship be builtin the United States and that its owners are U.S. citizens. See SCHOENBUAM, supra note XXX, at 499-501.

133 See SCHOENBAUM, supra note XXX, at 503.

134 A cargo reservation policy requires that some portion of the particular country’s exportsbe carried by its own domestic flag carriers. See FINK ET AL., supra note XXX.

135 See FINK ET AL., supra note XXX.

136 See Davies, supra, at 417. As Davies notes, liner shipping, especially since the advent ofcontainerization, demands a greater capital stake than other shipping sectors, because containerizedcargo vessels are by far the most expensive to build. See id. Moreover, once a schedule has beenagreed upon, cost items such as fuel, crew wages, subsistence, maintenance and repair–which mightbe regarded as variable in other industries–now become fixed costs which cannot be avoided withinthe short run planning horizon. See id. at 418. The only costs typical in general cargo ocean shippingwhich are truly variable are commissions paid to agents who secure cargo and actual handling costs.See id.

31

carriage to U.S. flag vessels132 and through a variety of cargo preferences, requiringthat most or all of various kinds of government impelled cargo be carried on U.S. flagvessels.133

Maritime nations have employed a variety of other protections for domesticcarriers, but there is reason to believe they exert less distorting effect on shippingmarkets. For example, while cargo-reservation policies for domestic flag vesselsremain prevalent,134 their influence is widely thought to have waned and a 2001econometric study by World Bank researchers found that such reservation policiesas still exist have little effect on pricing in the inbound U.S. shipping lanes.135

B. Cost Peculiarities of Liner Shipping. Even without market distortions caused by government subsidies and

protectionism, liner shipping would face special cost and other problems. First, it iswidely agreed that liner shipping of general cargo is beset by unusually high fixedcosts. It is estimated that in liner shipping of general cargo from 65 to 90% of allcosts are fixed. This problem arises both from the cost of contemporary containervessels and from the committed nature of scheduled transport services, which rendersmany operating costs invariant in the short run.136 Moreover, the maintenance ofscheduled service at typically demanded frequencies requires not one ship, but a fleetof them, plus all the appurtenant equipment and shore-based capital infrastructurethey require (containers, cranes, etc.). Thus the initial capital investment is

137 See Davies, supra, at 418-19. Davies calculated that the minimum capital outlay toestablish a new entry in the U.S.-Far East trade, where a minimum fleet size was estimated to be fivevessels, would be on the order of $374 million in 1978 dollars, even ignoring completely the costs ofmaintaining the shore-based administrative support such a fleet would require.

138 See Deltas et al., supra note XXX, at 4.

139 In the U.S. this would include notably wage rates of U.S. officers and crew, theemployment of which is mandatory on U.S.-flag vessels. See supra note XXX and accompanying text.

140 See Davies, supra note XXX, at 418-19.

141 See HERMAN, supra note XXX, at 29-30.

142 See S. Rep. 105-61, 105th Cong., 1st Sess., 1-2 (1997) (noting the industry’s “chroniccarrier conditions of overcapacity.”); H.R. Rep. 98-53(II), 98th Cong., 1st Sess., 5 (1984) (“Almostall studies agree that the industry is suffering from overcapacity.”).

32

immense.137 From the very beginning a key difference between liner operations andtramp shipping was the significant increase in fixed costs arising not only fromphysical investment in the vessels but also from the port-side investments necessaryfor the maintenance of regular liner operations.138

Moreover, liner shipping entails a large complement of “avoidable fixed”costs or “non-cargo” costs – that is, costs that do no vary with the quantity of cargocarried, but that are not incurred until a voyage is embarked upon. Most of thesecosts are outside the control of ship owners, and some of them are the subject ofregulatory price controls,139 and therefore carriers face significant limitations in theirability to cut operational costs even when they do so under pressure of pricecompetition. These include most obviously the substantial administrative andmarketing organization a liner fleet requires, which, given the committed nature ofscheduled transport services, is largely fixed in the short run,140 as well asoperational costs that are committed as soon as sailings are scheduled, including fuel,insurance, crew wages and costs, and maintenance.141

C. Capacity The liner industry has long perceived its chief obstacle to be a propensity

toward overcapacity, and there is little doubt that throughout its history the industryhas faced significant overcapacity problems.142 Overcapacity may follow in part fromhigh fixed costs. Large fixed costs naturally entail significant returns to scale, andas a result a more capacious ship is normally more efficient. However, the fact thatliner service by definition requires ships to sail regardless how much unused capacitythey may contain creates a perpetual risk of underutilization. Moreover, ships aredurable and exceptionally long-lived assets, and so even as more modern vessels

143 See ADVISORY COMMISSION REPORT, supra note XXX, at 68.


145 Indeed, a particular carrier’s cost functions are peculiar – namely, average cost constantlydecreases over the entire capacity of a particular ship, but capacity beyond full utilization of a givenship causes a sharp spike in average cost as the costs of an additional sailing are incurred, but thendecrease constantly across the full capacity of the added ship.

146 See ADVISORY COMMISSION REPORT, supra note XXX, at 68. Once a carrier has gone tothe expense of buying, crewing, and fueling a ship, the marginal cost of carrying an additionalcontainer on that ship is extremely low. Carriers may find it beneficial to accept cargo at any rateequal to or greater than this marginal cost whenever their ships are not completely full, even thoughthe rate may be insufficient to cover overhead. By taking this cargo, they utilize space that wouldotherwise be wasted, and the realize revenue that will at least cover marginal cost, but acceptance ofcargo on these terms can lead to ruinous rate wards because other carriers may follow suit in order tofill their own unused capacity, and rates may then spiral downward below those needed to produce areturn on investment. See id.

147 See HERMAN, supra note XXX, at 30-31.

148 See Davies, supra, at 432 (noting that demand for outbound shipping from New Zealandvaries considerably by season owing to that country’s large meat exports).

33

enter the market, older vessels may be resold at distress prices and remain inservice.143 Finally, the industry faces periodic cycles of slack and/or asymmetricaldemand, which are driven by inevitable shifts in currency fluctuations andinternational trade imbalances,144 and thus faces particular problems of forecastingneeded capacity.

Moreover, while overcapacity may be manageable in some industries, it is aparticular problem in liner shipping. Because a carrier’s costs are predominantlyfixed, the marginal cost of exploiting unused capacity within each particular ship isvery low,145 and therefore overcapacity can and does result in rate war.146 Costproblems have become only more acute with the advent of containerization and theincreased scale economies thereby engendered.

D. Unstable DemandFinally, these problems are compounded by the fact that demand for ocean

shipping is unstable. Because demand is derivative of demand for the goods to beshipped, it varies according to currency fluctuations and changing trade imbalancesgenerally.147 Demand for shipping also varies in some trade lanes as a consequenceof seasonal variations in outputs of particular national commodities, particularlyagricultural commodities,148 and because of institutional factors such as import

149 See Clinton H. Whitehurst, Jr., The Merchant Marine Act of 1936: An OperationalSubsidy in Retrospect, 1 J. L. & ECON. 223 (1958).

150 See FMC, 2002 ANNUAL REPORT, supra note XXX, at 23-24.

151 See FMC, OSRA REPORT, supra note XXX , at 12-13.

152 The Advisory Commission noted eight arguments frequently cited by proponents ofshipping conferences: “(1) chronic overcapacity which consequently leads to ‘destructivecompetition;’ (2) extreme rate fluctuations; (3) industry susceptibility to oligopoly behavior; (4) therequirements of international comity; (5) the need for common business practices in an internationalenvironment; (6) the presence of subsidized foreign carriers in the market; (7) the need to ensure thehealth of the U.S.-flag fleet; and (8) the desirability of achieving the efficiency benefits of jointactivities.” ADVISORY COMMISSION REPORT, supra note XXX, at iii-iv.


34

quotas, expert subsidies, cargo preference rules, and so on.149 But the mostsignificant problem is that during times of trade imbalance a carrier cannot ensure thesame degree of capacity utilization in both directions of a voyage, and therefore willsuffer less efficient operation in one direction.150 In fact, U.S. carriers have facedslack demand for outbound services for several years, and the imbalance has growneach year since 1995.151

IV. DESIRABILITY OF THE EXEMPTION:THEORY AND EMPIRICAL EVIDENCE

Arguments in support of the antitrust exemption take a variety of forms,152 buttheir gravamen is that special cost and capacity problems of liner shipping make itimpossible for the industry to arrive at efficient levels of supply, and that unbridledcompetition will lead to “destructive competition,” instability of prices, andundesirable oligopoly. These arguments rest almost exclusively on theoreticalmodels, as indeed was necessarily the case prior to OSRA, since the industry has onlyever faced actual competition for a period of a few decades in the 19th century andfor the seven years since OSRA. The recent evolution of the industry, in the wakeof OSRA’s first steps toward deregulation, shed empirical light on these matters, andhas brought the industry’s historical claims into doubt.

A. Capacity Rationalization and Unstable Price The industry’s chief argument for the conference system has always been that,

due to special economic dysfunctions, competition will not lead to efficient levels ofcapacity in liner shipping. There is no genuine doubt that the industry has sufferedfrom overcapacity throughout its history,153 but there are competing explanations why

154 See supra notes XXX and accompanying text; ADVISORY COMMISSION REPORT, supra,at 58; OECD, FINAL REPORT, supra note XXX, at 71 (“[T]here are many underlying reasons for . . .overcapacity, including state support for shipbuidling leading to exceptionally low . . . costs fornewbuildings.”).

155 See infra notes XXX and accompanying text (discussing evidence suggesting that theywere).

156 See infra notes XXX and accompanying text (discussing evidence suggesting that theywere).

157 See GAO REPORT, supra note XXX, at ii-iii. Devanney et al. found empirical support thatopen conferences – which, unlike closed conferences, have difficulty controlling capacity – lead toinefficient service competition. See Devanney et al., supra note XXX, at 162 & n.8.

158 See id.

159 See ADVISORY COMMISSION REPORT, supra note XXX, at 69; GAO REPORT, supra noteXXX, at v.

35

that should be.It may be that organic features of liner shipping lead to capacity problems.

However, there are also reasons to believe that capacity problems are driven by thecarriers’ own behavior, that exposure to competition will not increase thoseproblems, and that in any event antitrust immunity is not the best means to controlthem. First, at least some of whatever overcapacity exists is attributable not toendogenous market phenomena but to subsidization of shipyards and preferentialtreatment of national-flag carriers by many countries.154 Second, there is reason tobelieve that the carriers themselves have deliberately contributed to capacityproblems through the inefficient service competition typical of regulated or price-stabilized industries. If in fact conferences were able to maintain supra-competitiveprices under the conference system,155 and were able to constrain internal cheatingand lower-price entry,156 then conference members have little basis on which tocompete with one another except through improved service which, as a practicalmatter, means either more ships or more frequent salings.157 Increased capacity, otherthings being equal, necessarily means higher costs due to lower capacityutilization.158 Conference members, however, can still recoup the attendant loss ifthey are able to charge artificially inflated rates.

Carriers have also argued that without collective rate setting “destructive”competition will lead to unstable prices. First, while some shipper customers havealso expressed such a fear,159 it is not obvious that price volatility would beinefficient. Fluctuating prices are characteristic of many industries that are subject

160 See ADVISORY COMMISSION REPORT, supra note XXX, at 69.

161 See OECD, FINAL REPORT, supra note XXX, at 44-45 (noting that, according to shipperresponses to an OECD survey, annual rate changes averaged 5-10% in most trades, with 30% changesnot uncommon and some changes as high as 200%).


163 See ADVISORY COMMISSION REPORT, supra note XX, at 70.

164 Specifically, empty cores are said to be possible in markets which have some or all of thefollowing characteristics: (1) uncertain demand, (2) scale economies in production, (3) avoidablesupply costs, (4) products that cannot be stored cheaply, (5) fixed firm capacities, and (6) firmcapacities that are large relative to demand. See Wiley, supra note XXX, at 565.

36

to antitrust,160 and in liner shipping they might simply reflect sensitivity to changingsupply and demand. Moreover, there is some evidence that the conference systemactually promoted rate instability, especially in trades where there was lesscompetition.161 Finally, to the extent that fluctuating prices have been a problem,they are attributable in part to the traditional common tariff system itself, underwhich rates typically could be changed by the conferences unilaterally. Under servicecontracting, which has largely replaced the tariff system,162 rates are typically fixedfor specified periods. Long-term contracts are a well-known means by which toprotect against price instability and are used in a variety of industries.163

B. The Theory of the Empty CoreThe antitrust exemption’s academic proponents have applied a theoretical re-

tooling of the basic problem of capacity rationalization, known as the “theory of theempty core.” In principle the argument is simple, and if there are industries in whichempty cores can exist the liner shipping industry would be a good candidate. Theargument is that special cost or technological problems in some markets make itimpossible for competition to produce a stable long-run equilibrium price.164 Amarket has a “core” if there is a set of transactions between buyers and sellers suchthat there are no other transactions which can make some of the buyers or sellersbetter off. A basic implication of microeconomics is that such a “core” will survivein a competitive market where all firms are making zero economic profits. In amarket where the core is empty, no coalition of firms will be able to persist at zeroprofit; some firm will always eventually earn a surplus and thereby attract entry, butbecause the core is empty the new entry will result in all firms suffering losses.Likewise, because the core is empty, when firms exit due to economic losses, the

165 Empty core theory was first derived in the abstract by economist Lester Telser of theUniversity of Chicago and later applied (frequently by Telser’s own doctoral students) to a variety ofpractical situations, including liner shipping. The rendering of the argument presented in the text istaken from the excellent and straightforward explanation in the OECD report. See OECD, FINALREPORT supra note XXX, at 61-62.

For more formal theoretical explanations, including several attempted applications to linershipping, see LESTER G. TELSER, ECONOMIC THEORY AND THE CORE (1978); Varouj A. Aivazian &Jeffrey L. Callen, The Coase Theorem and the Empty Core, 24 J. L. & ECON. 175 (1981); GeorgeBittlingmayer, Did Antitrust Policy Cause the Great Merger Wave?, 28 J. L. & ECON. 77 (1985);George Bittlingmayer, Decreasing Average Cost and Competition: A New Look at the Addyston PipeCase, 25 J. L. & ECON. 201 (1982); Stephen C. Pirrong, An Application of Core Theory to the Analysisof Ocean Shipping Markets, 35 J. L. & ECON. 89 (1992); Abigail McWilliams, Rethinking HorizontalMarket Restrictions: In Defense of Cooperation in Empty Core Markets, 30 Q. REV. ECON. & BUS.3 (1990); William Sjostrom, Antitrust Immunity for Shipping Conferences: An Empty Core Approach,ANTITRUST BULL., Summer 1993, at 419; Lester G. Telser, Competition and the Core, 104 J. POL.ECON. 85 (1996); Lester G. Telser, The Usefulness of Core Theory in Economics, 8 J. ECON. PERSP.151 (1994); John S. Wiley, Antitrust and Core Theory, 54 U. CHI. L. REV. 556 (1987).

166 As Professor Sjostrom has observed, empty cores can stem from the lack of a price on theindustry supply curve for every possible quantity, which could occur because supply curves are“lumpy” or non-continuous. Where this is so, there are “gaps” on the supply curve into which the onlypossible equilibrium price might fall, thus making the core empty. Such a situation is arguablycharacteristic of liner markets if capacity can be added only in non-incremental blocs. See Sjostrom,supra note XXX, at 1166.

167 Suppose that a particular trade is such that when two ships service the route, the marketprice is above average cost, while when three ships service the route the market price is below averagecost. Suppose also that three different carriers want to serve this route. Since demand is such that onlytwo carriers can survive in the market, one firm will always be left out. If the other firms are makingprofit, the firm that is left out could seek to negotiate a deal with the customers of the other carriers,disrupting the original arrangement. See OECD, FINAL REPORT, supra note XXX, at 61-62.

168 In particular, William Sjostrom found some empirical evidence of conference behaviorarguably consistent with attempts to rationalize performance under empty core conditions. Sjostrom,however, relied on a highly simplified model of liner markets, see Sjostrom, Collusion, supra noteXXX, at 1162-70, used admittedly problematic data, and found no more than that “[t]he results [of

37

remaining firms will again be able to earn greater than zero profit.165 Liner shipping markets arguably contain empty cores. Because the entry of

even one new ship expands capacity not just incrementally but by the entire capacityof that new ship, existing firms earning a surplus will attract new entry that mayautomatically lead to overcapacity,166 and therefore to likely economic losses andpotential rate war. But as soon as a firm removes its bloc of non-incrementalcapacity, remaining firms may again be able to earn excess profit, again attractingentry and perhaps overcapacity.167 Several scholars have argued as much, and thereis also some empirical evidence in support of an empty core in liner shippingmarkets, though the evidence that exists is limited and has been strongly criticized.168

econometric analysis], although certainly not definitive, offer further evidence for the proposition thatmarket arrangements that appear to be cartels may be attempts to solve the problem of the emptycore.” Id., at 1177. Likewise, Stephen Pirrong argued energetically for the empty core hypothesis,but on little more than his econometric estimation of the cost function of one liner operator and theasserted longevity and universality of the conference system. See Pirrong, supra note XXX, at 107,116-29; see generally OECD, FINAL REPORT, supra note XXX, at 62 (providing criticism).

169 And, incidentally, if ever a real market were at risk of an empty core it should be one likeliner shipping, in which fixed and avoidable costs are extremely high and capacity was traditionallyadded only in large blocs. That fact that may shed on the general explanatory power of the theory.

170 Incidentally, another criticism exists of empty core theory concerning its feasibility inpractice, though the context of a regulated industry like liner shipping is one in which the criticism isprobably inapt. John Wiley has argued that core theory arguments should be ignored in antitrustpolicy because policies based on them will be difficult to administer. See Wiley, supra note XXX,at 571-72. In particular, he says that it will be difficult to distinguish them from badly-motivatedcartels and that even in industries in which the cartel addresses genuine empty core problems, thecartel could cause significant other injuries that would be difficult to distinguish from legitimate

38

Empty core theory is problematic for several reasons, and the experience ofthe liner shipping industry in the several years since OSRA suggests that, if they existanywhere, empty cores do not characterize liner shipping.169 As a theoretical matter,the model requires certain problematic assumptions. First, at least as applied to anextremely capital-intensive industry like liner shipping, it adopts an improbableassumption about the long-range investment planning by suppliers. It assumes thatwherever pricing above average cost poses short term gains upon entry an outsidefirm will enter, even though in an industry like liner shipping each new entry mightwell result in long run overcapacity requiring exit and potentially spurring rate war,entailing costs that could drastically outweigh short-run gains. What is important toa potential entrant is not the existing market price but the market price post-entry, andif a firm can foresee that its entry would force price below average cost it will notenter.

Empty core arguments also rely on a simplifying assumption about industrialorganization – namely, they assume that additional capacity can be added only innon-incremental blocs by autonomous and self-contained “firms,” which cancooperate by no means except naked restrictions on price or output. As applied toliner shipping the argument assumes that carriers can seek entry in markets withsurplus profits only by making irrevocable short term commitments of the fullcapacity of their own ships. This is at odds with experience, however; almostimmediately after OSRA the industry began reorganizing itself according to vesselsharing, space sharing, and other non-price operational agreements that allowedcarriers to rationalize capacity without naked price or output constraints. Thus,empty core theory ignores a solution that to the industry was made clear by the merepressure of competition.170 Incidentally, even were the Shipping Act antitrust

capacity-coordination efforts. See id. In other words, Wiley’s concern relates to the difficulty courtswould face in attempting to judge horizontal output constraints as empty core solutions in cases ofotherwise unregulated cartels. His criticism is probably not as compelling in the case of oceanshipping, because that industry has what is lacking in Wiley’s examples–a central administrator, theFMC, which oversees cartel behavior according to substantive guidelines and with investigative andenforcement powers.

171 It is unlikely that the typical bilateral or even multiparty operational agreement amongcarriers would violate antitrust, at least so long as it contains no direct constraints on price or output.See 13 HERBERT HOVENKAMP, ANTITRUST LAW ¶2100g (1999); U.S. Dept. of Justice and FederalTrade Commission, Antitrust Guidelines for Collaborations Among Competitors §3.3 (2000). Forwhat it is worth, such agreements would also likely qualify for treatment under the NationalCooperative Research and Production Act, 15 U.S.C. §4301-05, permitting them to avoid even thepossibility of treble damages liability upon the filing of a preliminary notice with the FTC and theAntitrust Division. See 15 U.S.C. §4305; 13 HOVENKAMP, supra, at ¶2100h.

172 Wiley, supra note XXX, at 575-76.

173 Even Lester Telser, the father of empty core theory, thought that long-term contracting canbe a solution to excess capacity where costs are lumpy and demand is uncertain. See Wiley, supranote XXX, at 565 (citing Lester Telser, Cooperation, Competition, and Efficiency, 28 J. L. & ECON.271, 274-76, 277-78, 284-85 (1985)).

39

exemption unavailable, procompetitive activity under operational agreements wouldlikely be legal under current antitrust law.171

A final point might be that even if liner shipping markets have empty cores,private capacity regulation of the kind envisioned by the conference system –regulation by self-interested market participants themselves – would be a poor meansto address the problem. As Professor Wiley put it, “[c]ore quota managers will findtheir powers for good tempt them to evil. They must be either saintly or regulated.. . . Legalizing core quotas would render useless the easy ways of outlawing cartels,because no simple, surefire test distinguishes laudable core management frominjurious cartel conduct.”172 A better solution would likely be long term contractingdirectly between shippers and carriers, a solution now freely available underOSRA.173

C. Susceptibility to Oligopoly ConditionsThe conference system’s defenders also argue that competition will lead to

concentration in the industry. The argument implies that oligopoly would benegative because it would facilitate supracompetitive price and consolidate controlover access to ocean transport. The prediction of concentration has strong theoreticalsupport – liner shipping is characterized by factors traditionally thought to facilitate

174 For example, ocean shipping theoretically should be susceptible to oligopoly pricingbehavior, because its customers are numerous, unaffiliated, and their identities change frequently.George Stigler has shown that oligopoly should be more stable where cheating is more difficult todetect, and that detection should be more difficult under these circumstances. See George J. Stigler,A Theory of Oligopoly, 72 J. POL. ECON. 44, 47 (1964). Likewise, liner markets may contain someentry barriers. See infra notes XXX and accompanying text.

175 See supra note XXX and accompanying text (discussing post-OSRA concentration).


177 A basic prediction of economic theory is that a collusive price-fixing cartel would seekto increase price: assuming a perfect capacity to constrain cheating and freedom from disciplinaryentry, the profit-maximizing cartel should act in the same manner as a single-firm monopolist. For afundamental statement on this point, see Stigler, supra note XXX.

178 See, e.g., Testimony of Christopher Koch, President, World Shipping Council, Before theHouse Committee on the Judiciary, International Liner Shipping Regulatory Policy and H.R. 1253,at 9 (2002); see generally OECD, FINAL REPORT, supra note XXX, at 30.

179 See OECD, FINAL REPORT, supra note XXX, at 31; Pirrong, supra note XXX, at 107.

40

oligopoly174 – and post-OSRA practical experience bears it out.175

But the relevant policy question is not necessarily whether competition wouldlead to oligopoly, but rather whether oligopoly pricing regulated by antitrust lawwould be worse than conference price-fixing, particularly price-fixing of the varietypracticed by the shipping conferences, protected as it was by the FMC as cartelenforcer and by the conferences’ own disciplinary procedures. In fact, there aretheoretical reasons to believe that capacity rationalization by oligopolists, constrainedby a competitive fringe, is to be preferred. A conference has an incentive to price offits least efficient member, and there is empirical evidence that they traditionally didso.176 An oligopolist, by contrast, is motivated by competition both from otheroligopolists and from fringe competitors to minimize costs. Therefore, competitionshould result in net pro-competitive consolidation, whereas conference price-fixingleads to subsidization of inefficient carriers.

D. Does the Conference System Result in Supracompetitive Price? If shipping conferences are harmful it would be chiefly because they charge

inefficiently high rates.177 Carriers, however, point out that rates have actually fallenfor roughly the past twenty years, and cite this as evidence that conferences have nopower over rates.178 Evaluating this claim is more difficult than it may seem becausecarrier cost data are hard to secure179 and because much of the evidence of price

180 The data are contradictory because with some frequency prices have behaved differentlyin seemingly similarly situated trades and they are difficult to interpret because of the size andcomplexity of the industry and the range of factors other than carrier market power that couldconceivably affect rates. See OECD, FINAL REPORT, supra note XXX, at 40.

181 Compare CLYDE & REITZES, supra note XXX (finding no correlation between marketshare and rates), with Fox, supra note XXX (finding such a correlation).

182 See CLYDE & REITZES, supra note XXX, at 2.

183 As the OECD Report pointed out, even the U.S. telephone industry experienced steeplyfalling prices for long distance service throughout the entire period up to the break-up of AT&T in1980, before which AT&T was a monopolist. See OECD, FINAL REPORT, supra note XXX, at 42.


185 That the influence of the conferences has declined appears now fairly well documented.The FMC has found that since OSRA the ability of conferences and discussion agreements to increaserates by way of voluntary pricing guidelines and non-binding common tariffs is dependent on demand. See FMC, OSRA REPORT, supra note XXX, at 14.

41

behavior is contradictory and hard to interpret.180 However, the evidence that existsis either inconclusive or tends to suggest that conferences had some ability to inflateprice.

The evidence is in conflict as to whether there was ever a correlation betweenconference market share and freight rates.181 However, a 1995 econometric studyby Federal Trade Commission staff economists found that freight rates went downsignificantly where carriers were permitted directly to negotiate independent servicecontracts.182

Moreover, the fact that price has fallen over a given period is as theoreticallyconsistent with market power as it is with lack of market power, since evenmonopoly maximizing rates are sensitive to changes in demand over time.183 Indeed,for what it is worth, the practical evidence is more consistent with market power,because though rates have been in steady decline for some time, the decline beganonly at about the time of the first U.S. deregulatory reforms in 1984, and havedeclined from a peak during the 1970s.184 This steady decline, then, coincides witha steady decrease in the influence of the conferences themselves,185 and if anythingis more consistent with the idea that conferences did exert some market power.

E. Will Competition Lead to Inadequate Returns and Lower Service Quality? Carriers have argued that open competition would lead to desperate

overcapacity and pricing below cost, and in support they frequently point out thattheir industry already performs poorly economically. However, while liner shippinghas performed comparatively poorly as among industries generally, the evidence that

186 See OECD, FINAL REPORT, supra note XXX, at 45-46.


188 See GAO REPORT, supra note XXX, at 11-17; H.R. Rep. 98-53(II), 98th Cong., 1st Sess.,4 (1984) (House Committee Report accompanying 1984 Act, noting reliance on GAO findings).

189 See OECD, FINAL REPORT, supra note XXX, at 47, 53.

190 See, e.g., DEAKIN & SEWARD, supra note XXX.


192 See OECD, FINAL REPORT, supra note XXX, at 22, citing FEDERAL MARITIMECOMMISSION, SECTION 18 REPORT ON THE SHIPPING ACT OF 1984 (1989).

42

exists suggests that it performs about as well as other transportation sectors.Moreover, the best-performing liner carriers perform quite well by comparison torailroads and other transport industries, and most of the top 20 carriers have been inbusiness for over 20 years – that is, throughout the entire period of price competitionunder deregulation.186 Thus, while it may be that many carriers have struggled sinceprice competition began, that may reflect no more than the healthy market functionof forcing exit of higher cost firms. This possibility is supported by the massiverecent concentration in the industry.187 Indeed, the early deregulatory steps appearingin the Shipping Act in 1984 were taken in part because a 1982 report of the GeneralAccounting Office found the industry to be doing much better than it claimed.188

In any case, it appears that the carriers may have contributed to whateverunderperformance they have experienced by engaging in service competition throughinvestment in overcapacity. As already discussed, the conference system at leastsometimes may have encouraged inefficient service competition, which would befinanced through profit.189 There is empirical evidence to the contrary, finding nolink between conference control and overcapacity,190 but the evidence that exists isbased only on study of closed conferences. Closed conferences, by hypothesis,should be better able to rationalize capacity. While closed conferences are stilltheoretically permitted in some trades, they have long been illegal in U.S. shippingand in practice even where they are permitted they rarely exercise membershiprestrictions and operate essentially as open conferences.191

Carriers also argue that the lack of adequate return they anticipate undercompetition and the capacity instability it will cause will result in a loss of servicequality. A generalized decrease in shipping rates has been ongoing since theenactment of the Shipping Act in 1984, with its initial liberalization of individualservice contracting, and yet there is no evidence of an impact on service.192

193 A basic prediction of economic theory is that in the absence of entry barriers any abuseby a cartel of its position should create the possibility of surplus for other competitors, and thereforeshould invite both disciplinary competition and opportunistic cheating by its own members. SeeFREDERIC M. SCHERER, INDUSTRIAL MARKET STRUCTURE AND ECONOMIC PERFORMANCE 171 (2d ed.1980) (noting “the tendency for informal price-fixing and output-restricting agreements to breakdown.”)..

194 Ships, obviously enough, are “moveable” capital, and it is therefore often suggested thatparticular trade routes are highly contestable. See HERMAN, supra note XXX, at 5; Sjostrom,Modeling Competition, supra, at 2

195 See John Davies, Impediments to Contestability in Liner Markets, 25 LOGISTICS &TRANSP. REV. 325 [DATE]

196 See, e.g., Pirrong, supra note XXX, at 116; Sjostrom, Modeling Competition, supra noteXXX, at 2. Professor Sjostrom has gone so far as to say that the conference system is popular withshipping consumers, though his assertion seems somewhat incautious. The evidence he cites comesmainly from small shippers, see Sjostrom, Modeling Competition, supra note XXX, at 2 & nn.7-8,who would be less able to negotiate favorable rates with carriers even in competitive environments,and it is also at odds with the large body of evidence from shippers of all sizes. See ALEXANDERREPORT, supra note XXX, at 304-08 (noting large volume of complaints Committee received fromshippers and the broad shipper support for extensive regulation of conferences, including strongsupervision of rates); see also McGee, supra note XXX, at 238-39.

43

V. WHY WOULDN’T HARMFUL CONFERENCE ACTIVITY INVITE CHEATING AND NEW ENTRY?

A final problem remains: If conferences are inefficient, and theircollaborative work is not merely a procompetitive effort to control otherwiseunmanageable capacity, then abuses by them should invite cheating and new entryand render them comparatively harmless.193 Moreover, carriers note that individualships by their nature are moveable capital194 and predatory retaliation against newentrants is believed by some to be slow.195 Carriers and some academics argue thatliner conferences have historically been long-lived and stable, a result that should notobtain if the conferences have ever abused their position.196

These claims are subject to several theoretical criticisms and are at odds withrecent experience under OSRA’s deregulatory innovations. Most importantly, unlikecartels operating under normal competition and subject to antitrust law, linerconferences throughout their history have had the benefit of a powerful, government-sanctioned cartel enforcer – the FMC and its predecessors. Prior to OSRA bothconference tariff rates and deviations from them were required to be on file with the

197 See supra note XXX. Even under the 1984 Act, direct negotiation between shippers andconference members was limited by FMC rule. See Paul S. Edelman, The Ocean Shipping ReformAct of 1998, 9 CURRENTS 65, 65 (2000).

198 Enforcement of tariff terms, including prosecution of secret rebates and other undercuttingactivities, was added by the 1961 Shipping Act amendments. Such enforcement grew to become adominant feature of the FMC’s activities – perhaps because the agency was incentivized by the largepenalties available in tariff enforcement – and by 1992 they constituted two-thirds of the agency’senforcement activities. See ADVISORY COMMISSION REPORT, supra note XXX, at 107-10. Tariffenforcement was effectively ended by OSRA, with the adoption of freely available service contracting.Cf. 46 U.S.C. App. §1709(b)(2)(A) (prohibiting the charging of rates by a common carrier notcontained in a filed tariff or service contract).

Enforcement of tariff rates was urged by the Alexander Committee as a means of protectingshippers, given the Committee’s concern for the perceived harm of discriminatory pricing. SeeADVISORY COMMISSION REPORT, supra note XXX, at 14. Presumably neither the Committee nor theCongress of 1916 intended this provision as a key protection for conference market power, but thatis arguably what it has been.

199 For example, conferences in most trades outside of U.S. foreign commerce have alwaysbeen “closed,” meaning they simply restricted their own membership, see ADVISORY COMMISSIONREPORT, supra note XXX, at 37; GAO REPORT, supra note XXX, at 4, and often controlled their owncapacity by limiting sailings by individual members, see id. To suppress internal price competition,conferences employed revenue pools, under which revenues earned by individual member companiesserving the same ports were pooled together and periodically distributed among the membercompanies according to pre-approved schedule, thus eliminating the incentive for secret undercutting.See GAO REPORT, supra note XXX, at 4-5. Prior to 1984 revenue pools required FMC priorapproval, but since that time they have been permissible and within the antitrust exemption, so longas filed and not contrary to the standard of §1705(g). As for price competition by independents,conferences seem mainly to have employed various species of “loyalty contracts” with shippers, seeinfra notes XXX and accompanying text, and service competition, such as better sailing schedulesthrough capacity rationalization that independents could not match. See GAO REPORT, supra noteXXX, at 5. Finally, conferences sometimes resorted to more drastic measures, such as boycotts ofshippers who used non-conference carriers and the so-called “fighting ship,” a vessel subsidized byconference members that would meet or undercut the lower rates changed by any independent until

44

FMC and available to public review,197 and between 1961 and 1999 they wereenforced by the FMC as a matter of law.198 Conference members had no legal rightof independent action prior to OSRA, and, even where such action was permitted bya particular conference, shipping law prior to OSRA required that the rate and termsof any individual service contract be filed with the Commission and available topublic review. Thus, cheating by conference members was not only difficult andeasily disciplined by conferences, it was in fact illegal. In short, until OSRA the linerconferences had the one thing that most cartels lack, and the one thing that makesmost cartels unstable – a highly effective regime of cartel enforcement.

Second, prior to OSRA the conferences had a variety of devices at theirdisposal to discipline their markets and protect collectively set rates,199 and even

it either left the trade or joined the conference. See GAO REPORT, supra note XXX, at 5. Bothpractices have always been illegal under U.S. shipping law, as contrary to the basic “commoncarriage” philosophy of U.S. policy. Cf. 46 U.S.C. App. §1709(b)(3) (prohibiting retaliatory boycott);id. at §1709(b)(6) (prohibiting fighting ships).

See generally Sjostrom, Modeling Competition, supra note XXX, at 1 (“At different times,subject to various regulations, [conferences] have set tariffs, employing policing agencies to check onadherence to the tariff. Members have been fined out of the membership bonds they post. . . . Theymay also pool revenues and allocate particular ports”).

200 For example, U.S. law has long mandated “open conferences,” without membershiprestrictions, has always prohibited “fighting ships,” and has traditionally placed some limits on suchdevices as revenue pooling agreements and loyalty contracts. See GAO REPORT, supra note XXX,at 5-9.

201 And, not least important, they sometimes used devices that were not permitted by law. SeeH.R. Rep. 98-53(II), 98th Cong., 1st Sess., 4-5 (1984) (describing the late 1970s U.S. prosecution ofexecutives of North Atlantic Conference carriers for collusive conduct beyond the scope of approvedagreements, resulting in multi-million dollar fines).

202 Under a dual rate contract the shipper agrees to purchase exclusively from the conference,and in return receives a discount. If it then purchases from a different carrier, it must pay damages. Another type of loyalty contract is known as the “deferred rebate,” under which if a shipper purchasesexclusively from a carrier or conference for some stated period it accrues a rebate, but the rebate isreturned only at the end of that stated period. If the shipper purchases from another carrier during theperiod it forfeits the rebate. Though some have argued that the two arrangements have differentsubstantive consequences, it seems more probable that in the absence of enforcement costs they areidentical. See William Sjostrom, Monopoly Exclusion of Lower Cost Entry, 22 J. TRANSPORT ECON.& POL’Y 339, 339 (1988) [hereinafter “Sjostrom, Monopoly Exclusion;”]; John S. McGee, OceanFreight Rate Conferences and the American Merchant Marine , 27 U. CHI. L. REV. 191, 231-38(1960). In any event, deferred rebates were long illegal under U.S. law, while dual rate contracts weretraditionally permitted. See Sjostrom, Monopoly Exclusion, supra, at 339. As to the brief periodduring which dual rate contracts were also illegal under the Isbrandtsen case, see supra note XXX.

203 Professor Sjostrom argues that a component of the transportation cost faced by a shipperis the sum of the physical storage cost of inventory and the lost time value of the good while waitingunsold in storage. This cost can be lowered by paying a premium to a carrier to add greater frequencyof shipment, and, other things equal, a rational shipper minimizes transportation cost by paying foradded frequency until the frequency premium exceeds the recouped storage cost. In short, the demandfor frequency of shipping service is downward sloping. It follows that wherever the price of added

45

where some restrictive devices are prohibited, as under U.S. law,200 conferenceshistorically retained a variety of disciplinary tools.201 For example, U.S. policy hashistorically permitted the use of one particular form of entry deterrence that isthought to have some substantive effect even in the absence of other entry barriers– the so-called “dual rate” or “loyalty” contract.202 Even one of the conferencesystem’s defenders has argued that loyalty contracts can deter entry even by lowercost entrants, so long as they are capacity constrained.203 In any event, a chief

frequency falls, other things equal, shippers will desire greater frequency. Thus, the entry of a lower-cost new entrant might attract shippers. However, if the entrant

is capacity constrained it may be unable to satisfy the demand generated by any price lower than theincumbent’s cost. If so, shippers would lose some surplus. Shippers may be able to earn more overallsurplus by agreeing to an exclusive contract with the incumbent, under which the incumbent mightcharge more than the lower-cost entrant’s minimum price, but at which the incumbent would provideall needed demand. Shippers will prefer such a contract if: (1) the product of the additional capacitypurchased under the exclusive contract times the price at which shippers would have purchased thatcapacity exceeds (2) the lost surplus suffered by purchasing from the lower-price but capacity-constrained incumbent. Professor Sjostrom posits that incumbent conferences will normally be ableto satisfy excess capacity demands, whereas non-conference entrants will frequently be capacityconstrained. See William Sjostrom, Monopoly Exclusion of Lower Cost Entry, 22 J. TRANSPORTECON. & POL’Y 339, 341-42 (1988). See also Jong-Say Yong, Excluding Capacity-ConstrainedEntrants Through Exclusive Dealing: Theory and Application to Ocean Shipping, 44 J. INDUSTRIALECON. 115 (1996) (applying game theoretic model to reach similar result).

204 [GENERAL CITES]

205 [CITE ALEXANDER REPORT]; RING COMMISSION REPORT, supra note XXX, at 20-27.

206 This is so because virtually all general cargo shipping is now by container ship andmodification of other ships for container carriage would likely be prohibitively expensive.Containerships are much more expensive than ships fitted for other carriage–a containership can coston the order of three and a half times that of a similar sized bulk carrier. See Davies, supra note XXX,at 417; see also HERMAN, supra note XXX, at 6. Thus, entry would be difficult by ships speciallysuited to refrigerated cargoes, those designed for bulk shipping, and so-called “roll on/roll off” ships,which are designed to handle wheeled cargo trailers, containers with chassis, and self-propelledequipment, such as automobiles, which can be driven onto and off a vessel over ramps. See generallyADVISORY COMMITTEE REPORT, supra, at 17 & n.3.

207 See Sjostrom, Modeling Competition, supra note XXX, at 6.

46

characteristic of most conferences has been their well-funded, well-staffed andinfluential central secretariats,204 a feature that has been noted nearly as long asconferences have been known to exist.205

Moreover, despite superficial appearances there remains steady debateconcerning the contestability of liner markets, and in fact there is reason to believethat substantial portions of world shipping capacity are not suitable to competitionin the general cargo trades that make up most Shipping Act-covered commerce. Bulkcargo ships and some others cannot be switched into general cargo competition atlow cost,206 there is at least some specialization of ships to particular routes even ingeneral cargo trades,207 and at least some other non-trivial sunk costs may inhere in

208 See CLYDE & REITZES, supra, at 22 (hypothesizing that “there my be sunk costs involvedin serving a given route (i.e., costs or warehouses, cargo-handling equipment, and other terminalfacilities).”).

209 See OECD, FINAL REPORT, supra note XXX, at 20, citing MARY R. BROOKS, SEA CHANGEIN LINER SHIPPING: REGULATION AND MANAGERIAL DECISION-MAKING IN A GLOBAL INDUSTRY(2000).

210 See OECD FINAL REPORT, supra note XXX, at 62, citing N. Shashikumar, Competitionand Models of Market Structure in Liner Shipping, 15 TRANSPORT REV. 3 (1996).




47

serving individual routes in general cargo trades.208 Thus, except for high value andvery low value goods, the bulk of cargo transported by containerized ship cannot becost-effectively shipped by alternative means.209

In any case, the frequent suggestion that liner conferences are long-lived andstable is misleading in two respects. First, while it is true that throughout its historyand up until OSRA the industry was dominated by conferences, the conferencesthemselves typically lasted only a few years, and individual conference membershipfluctuated along with carriers’ business strategies.210 Second, OSRA’s first stepstowards deregulation and the introduction of price competition through confidential,individual service contracting have hastened the virtual demise of the conferencesystem in less than ten years.211

But finally, the most telling evidence may simply be that in the short periodsince OSRA, during which carriers were no longer able to fix binding commontariffs, the conference system essentially collapsed but the industry appears to haveperformed well.212

VI. CONCLUSIONS

A. Retention of the Antitrust ExemptionThe question remains whether the antitrust exemption that continues under

OSRA causes any negative consequences calling for further deregulation. In fact,though OSRA made genuine price competition possible for the first time, at least afew problematic behaviors persist that might be better treated were antitrust to apply.

First, the “discussion agreements” and “voluntary guidelines” for servicecontracting still tolerated by OSRA routinely involve a large amount of informationsharing that might violate U.S. antitrust were it not for the exemption,213 and it is not

214 See FMC, OSRA REPORT, supra note XXX, at 13-14, 28-29.


216 See OECD, FINAL REPORT, supra note XXX, at 43-44. These apparently collusively fixedsurcharges include a variety of costs that presumably should vary as among carriers, such as equipmentrepositioning charges and paper work filing. Carriers have also managed to pass on a number of majorvariable charges to shippers (such as currency and fuel price fluctuations), so that shippers are thenfaced with rates that vary highly from the published tariff. Finally, the lack of transparency involvedin the assessment of these charges and the fact that they are presented to shippers as non-negotiable“direct costs” suggest that surcharges are simply a continued price-fixing effort. See id.

217 See, e.g., Heaver, supra note XXX (finding commodity unit value most important factorin freight rates); Jansson & Shneerson, supra note XXX (same); McGee, supra note XXX, at 224; seealso Bryan, supra note XXX (finding price discrimination across all commodities). Note that whileU.S. shipping law long prohibited “discrimination” amongst shippers, it prohibited it only as among“similarly situated” shippers. Thus, conferences and carriers were free to set different rates fordifferent kinds of cargo. See ULLMAN, supra note XXX, at 50.

218 It has long been believed that oligopolistic pricing should be more effective where demandis inelastic. See McGee, supra note XXX, at 197.

48

obviously necessary to the functioning of the shipping market. In fact, it appears thatdespite the introduction of competition, this conduct has facilitated some collusion.Though the voluntary guidelines have proven vulnerable to independent servicecontracting, particularly in times of overcapacity, they do appear to facilitategeneralized rate increases in times of high demand and capacity utilization,214 andcarriers may benefit more from rate increases in times of high demand than shippersdo from rate troughs in times of low demand.215 More significantly, the guidelinesand discussion agreements very clearly have facilitated absolute price-fixing of themany ancillary “surcharges” that carriers pass on to shippers, even in independentservice contracts, notwithstanding that freight rates themselves remain negotiable.These charges often constitute significant portions of overall transport cost, and alsoresult in major (and sometimes unseen) shifting of risk to shippers.216 Finally, therehas always been evidence that carriers are able to price discriminate as betweenshippers of high-value and low-value goods, because elasticity of demand for oceanshipping is thought to vary indirectly with the value of the underlying cargo.217 Thus,especially in periods of slack demand, it stands to reason that carriers are able to useconferences and discussion agreements, in which they exchange price and customerinformation, to fix prices as to high value goods.218

Second, while OSRA has introduced price competition, independentlynegotiated rates are probably available only to shippers large enough to exertinfluence in negotiation with individual carriers. Whereas published tariff rates nowlargely serve as benchmark prices below which large shippers enjoy deep discounts

219See OECD, FINAL REPORT, supra note XXX, at 66; Papavizas, supra note XXX, at 489.


221 See HOVENKAMP, supra note XXX, at ¶2100b (“While many joint activities are clearlyanticompetitive and many others are clearly competitive, in the middle are a significant number whoseeffects are ambiguous, at least upon an initial look.”); OECD, FINAL REPORT, supra note XXX, at 27(noting problems with operational agreements on file).

222 See OECD, FINAL REPORT, supra note XXX, at 49 (“It is also telling that a growingproportion of the top 20 operators’ fleets is made up of time-chartered vessels, indicating a trend awayfrom self-ownership to relatively more flexible asset management arrangements.”); id. at 57 (“Slotchartering allows carriers to respond flexibly to demand without necessarily purchasing a newvessel.”).

49

through service contracts, small shippers typically have no choice but to accept thebenchmark rate.219 Thus, as to them, the conferences remain effective price fixers.

Furthermore, the operational agreements now prevalent amongst carrierscome within OSRA’s antitrust exemption, and while they appear to promiseprocompetitive benefits, there is no obvious reason they should not be subject toantitrust. Procompetitive aspects of inter-carrier operational agreements would likelyonly ever be subject to rule of reason analysis,220 and even if that may burden inter-carrier negotiations to some extent there is no obvious reason such agreements shouldbe treated differently than any other efficiency enhancing cooperative behavior.Moreover, while they appear likely to result in productive efficiencies, collaborationsamongst carriers can easily shield conduct harmful to competition.221

Opposed to these harms of unregulated collusive activity is the now freelyavailable access of the liner industry to means of rationalization that do not harmcompetition and need not be exempt from antitrust. Shippers and carriers can protectthemselves from rate instability, demand uncertainty, and underutilized capacitythrough long term service contracts. Through operational agreements, carriersthemselves now have access to means of more flexible asset-managementarrangements than outright ownership of non-incremental blocs of capacity, the entirecapacity of which must always be irrevocably committed to both legs of a particularroute in every case.222

B. The Bigger Picture One final question remains, and that is just how the congressional action of

1916 should be understood historiographically. It seems likely that when Congressenacted the Shipping Act it mistook the facts before it. The Alexander Committeeand the Congress of 1916 believed they were dealing with a new or special industrialphenomenon, and that the economic “crisis” in ocean shipping reflected general crisisin American industry. Very similar patterns played out in other transport sectors in

223 See JAMES WILLARD HURST, LAW AND SOCIAL PROCESS IN UNITED STATES HISTORY 28-92(1960) (arguing that American public policy is characterized by “drift” rather than “direction”); JOHNKINGDON, AGENDAS, ALTERNATIVES AND PUBLIC POLICIES (2d ed. 2002) (laying out theory of creationof policy “agendas”).

224 See, e.g., WORLD SHIPPING COUNCIL, supra note XXX.

50

the early twentieth century; indeed, this history seems part of the major rhetoricaleffort that culminated in the National Industrial Recovery Act fifteen years later.

A plausible alternative explanation might be this: what was before Congresswas not an irremediable crisis, but a period of readjustment in shipping markets tonew long term equilibrium, a readjustment that was cut short by congressionalsanction of shipping cartels. This readjustment was simply a part of generalreadjustment in American industry, a readjustment that was wrenching because it wasboth large and rapid.

A cynical view would be that the carriers, even though they knew there wasno crisis, perceived an opportunity for legal sanction of profitable anti-competitiveactivity – that they knew a good thing when they saw it. They knew their industrywas simply in readjustment, but sought to avoid the painful contest by which someof them would exit the industry. While it seems impossible that there was not at leastsome self-interest in these events, this version of the story is not entirely compelling.For one thing, it perceives theory is simply a tool of evil – a pliant Congress wassusceptible to an industry-friendly abstraction from events.

Rather, a reasonable read is that Congress misunderstood the situationbecause, at the time, virtually everyone misunderstood it. The misunderstanding wasin part a legacy of nineteenth century economic orthodoxy struggling to make senseof unprecedented circumstances. In addition to avoiding undue aspersions at oddswith Occam’s Razor, this interpretation has the added virtue that its description ofCongress is in character. Congress, like other legislatures, has a history of actionwithout overarching plan. Its behavior over time is well characterized as a series ofreactions to emergent events and perceived crises, few of which fit into any large planand most of which have long term consequences considered by essentially no one atthe time of their enactment.223

However, cynicism is not so easily avoided as to the shipping industry’scentury-long efforts to defend the exemption. In light of events of the past fewdecades, which ever more clearly show that the traditional explanation for theexemption was faulty, their continued arguments suggest that they are no longersimply believers in theory faced with difficult circumstances. Even to the present daythey persist in stressing the economic peril of competition, in order to avoid that finalbit of deregulation they most likely will suffer in the coming few years.224 In alllikelihood, continued exposure to competition will result in turmoil for many carriers,

51

just as they predict, but only because the near-century of U.S. shipping policycontributed to industrial organization, overcapacity, and other settled expectationsthat were in themselves inefficient.

the demise of ocean shipping regulation: a study in the evolution of ...

Documents

Transcript of the demise of ocean shipping regulation: a study in the evolution of ...