JPRI Working Paper No. 6:January 1995
Market Access in the Global Economy: The Problem of Cartels
by Alan Wm. Wolff

A few precautionary notes: I am speaking for myself, not for my firm or any client. My approach is that of an indomitable optimist as a trade negotiator, and thus represents a triumph of hope over experience.

Since the inception of the GATT in 1948, there have been eight major rounds of multilateral trade negotiations. These rounds have been directed almost entirely at reducing government-imposed barriers to the flow of goods and services in international commerce. By and large these negotiations have been successful, although not perfectly so. Tariffs have fallen dramatically, and government administered import quotas and other quantitative restrains have been greatly reduced. Government procurement has been at least partially opened to competitive bids from other countries. GATT members have developed codes governing the use of subsidies, product standards and, most recently, services and intellectual property. These accomplishments have made possible a dramatic expansion of international trade. Much of the original agenda of the founders of the GATT has now been achieved, and from that we can take great satisfaction.

But despite the GATT's many achievements, some large pieces of work remain to be addressed. A number of types of market barriers exist for which the GATT has not yet found any answers. Far and away the most important is the problem of restrictive private practices that inhibit market access in international trade. Indeed, it is here that the seeds of the biggest trade conflicts of the next decade are likely to be found.

This is not a new problem. From the beginning, the planners of the postwar economic order were concerned that the elimination of government barriers to trade could be nullified by the restrictive acts of private cartels. As the U.S. State Department noted in 1945 in a document prepared for the International Trade Organization (ITO):

Goods can surmount a tariff if they pay the duty: they can enter despite quota if they are within it. But when a private agreement divides the move between the zones while the contract is in force.

Let me illustrate this problem with a case study. My colleagues and I did not set out to uncover an international steel cartel last year. We accidentally tripped across one, so to speak, because we had a trade problem. Our firm was in the process of litigating antidumping and countervailing cases on behalf of U.S. producers of flat-rolled steel. During the litigation, a number of individuals involved in the trade came forward and indicated to us that the dumping problem was, to a considerable degree, a function of cartel arrangements abroad. We were aware of some cartels in steel--particularly those which operated with government approval in the EU and Japan--but we had no idea of the full extent to which cartels regulated international trade in this industry. The term "cartel" is used here in its classic sense--these arrangements are not about joint R&D, joint manufacturing, or some other similar kind of benign or beneficial joint activity. The purpose of these cartels is price-fixing, division and allocation of markets, and joint restraints on output and delivery--the kind of activities that are prohibited in most OECD countries.

First and perhaps most important is the arrangement between the integrated mills of the European Union, on the one hand, and those of Japan and Korea, on the other hand.

-First, steel competition within the EU and within Japan is limited by officially-sanctioned restraints on output which have one objective, that of maintaining price stability. Both the Commission and MITI publish quarterly forecasts of steel output that are intended to serve as production ceilings for each mill. These are euphemistically referred to as "market stabilization" measures.

-The European and Asian mills have taken the additional step of entering into the so-called London Agreement, or East of Burma Agreement, which divides the Eastern Hemisphere into spheres of influence, with Burma serving as the dividing line. East of Burma the Europeans are limited to very small export quantities--less than 400,000 tons annually--and the Japanese and South Koreans are similarly limited west of the line. Violations are punished by dumping of twice the tonnage in the home market of the offending firm, at a time and place that is least convenient. The London Agreement can be characterized as a sort of private bilateral agreement between two regional cartels, one in Europe and one in Northeast Asia.

-Tonnages and prices are adjusted in quarterly meetings. For example, in Kyoto, in April of 1994, it was agreed that the Europeans and Asians would raise world export prices for the third quarter of 1994. In March of 1993, a similar meeting was held in the Hotel Los Monteros in Marbella, on Spain's Costa del Sol.

-South Korean-Japan trade is similarly regulated by a mill-to-mill agreement that restricts South Korean exports to Japan to 2.2 million tons per year.

What effects have these arrangements had on trade? An economic study currently underway shows that steel trade patterns conform to the quotas set under the London Agreement. Chart 1 depicts five years' worth of shipments of a commodity grade of hot-rolled sheet by EU firms and Japanese firms to the 28 biggest export markets in the Eastern Hemisphere, outside of the EU and Japan themselves. The dotted line represents Burma, the demarcation line, with markets to the west shown to the left of the line, and the Pacific Rim markets on the right. The Japanese have almost all of the trade in their sphere and the EU mills enjoy similar dominance in theirs. The pattern is the same for commodity-grade cold-rolled sheet steel; in fact, it is the same for product after product in the flat-rolled categories.

Of course it could be argued that these trade patterns are a function of distance-related factors such as freight costs. But almost no freight cost differential exists, for example, between a Japanese shipment to Thailand and one to India. The same is true for European shipments headed in the other direction. If distance were a factor in export competitiveness, there would be a gradual tailing off of EU shipments as one moved further east, instead of the abrupt demarcation at the Burma line.

The cartel's restraining effect on exports can be illustrated in many other ways. Chart 2 shows Spanish exports of flat-rolled steel to Japan, compared with those of all other EC mills combined. In the mid-1980s, Spanish shipments dwarfed those from the rest of the Community--a fact that shows, by the way, that it is economical for European mills to export to Japan, and that the reason for low EU sales is not freight costs or some other logistical factor. But at the end of the 1980s the Spanish mills joined Eurofer, the Community's steel producers' association; and Spain's exports to Japan fell sharply. They are still dropping as Spain is integrated into the London Agreement. In light of Spain's experience, it will be interesting to observe how Sweden's and Austria's steel trade will be affected as these countries join the EU on January 1, 1995.

Another component of the international cartel regime is a so-called "non-aggression" agreement between flat-rolled steel producers in Japan and Canada. As it has been described to us, under this agreement the Canadians ship no flat-rolled steel to Japan, and the Japanese ship no flat-rolled steel to Canada, except for some specialized items needed by Japanese transplant firms in Canada. Both countries are large producers and consumers of these products, and Canada ships large volumes of flat-rolled steel to East Asian markets like South Korea and Taiwan, where prices are much lower than in Japan, but nothing--nothing--to Japan. Japan ships enormous tonnages of flat-rolled steel to the United States, where prices are generally lower than in Canada, but virtually nothing to the higher-priced Canadian market.

Finally, we learned that Japanese mills have reportedly negotiated a series of mill-to-mill agreements with various foreign producers to regulate the price and volume of steel moving in both directions between each of these countries and Japan (Chart 3). The Japanese producers agree to limit the volume of their exports to each country, and to avoid price reductions below a certain level, and in return a similar pledge is given with respect to the Japanese market. The enforcement of mill-to-mill agreements is based on the non-Japanese mills' fear of what is called the "total power" of the Japanese mills. The foreign mills are dependent upon the Japanese for technology and in some cases financial assistance and semifinished steel. The Japanese producers enjoy additional leverage because they buy large quantities of raw materials from some of the countries concerned, such as Australia, and can switch suppliers to show their displeasure. The Japanese mills have a large reserve of production capacity and can dump in the home market of any foreign mill that does not toe the line.

In all of these arrangements one country is conspicuous by its absence--the United States. For U.S. firms taking part in cartels like these is simply not an option, and the foreign mills reportedly take care not to discuss the American market in their deliberations. This is not because American businessmen are more virtuous or because foreign executives have no interest in regulating competition in our market. It is because U.S. antitrust enforcement is taken seriously by everybody, in the U.S. and abroad. But for the U.S. steel industry, the offshore cartels matter because they have fostered endemic dumping into the U.S. market which has kept the industry in a state of crisis for decades, depressing operating rates, profits, and ultimately investment levels, while strengthening its foreign competitors.

The nexus between cartels and dumping was recognized a hundred years ago, when dumping first emerged as a common trading practice, and it remains a basic fact of commercial life today. Indeed, some cartels have regarded dumping as so central to their success that they have developed institutional mechanisms for spreading the costs associated with dumping on an even basis to all members, through a levy imposed on the members. The rationale for dumping is simple:

-Most cartels seek to create an artificial shortage in the home market by restricting output, deliveries, or both in that market. That means surpluses must be removed from the market if prices are to be kept high.

-Surpluses are diverted from the cartel's home markets onto export markets at whatever price is needed to sell them--that is, they are dumped.

Over a long period of time cartels and dumping affect investment patterns. Excessive capacity is built up and sustained in cartelized markets, and dumping serves as a deterrent to investment in more open markets. That is what has happened in steel.

The U.S. steel industry has made considerable use of antidumping measures and has been severely criticized for it. Antidumping measures, it is charged, are protectionist and "disruptive." But one must ask, disruptive of what--the free flow of trade, or the clandestine arrangements of international cartels? In steel, at least, the latter is more nearly accurate. I am not alone in making this connection; for their part, cartel members have repeatedly blamed U.S. antidumping measures for foiling their market-regulating arrangements. When EU, Japanese and South Korean company representatives met in March 1993 in Marbella to discuss prices, market division arrangements, and other cartel business, one of their principal concerns was how to counteract the antidumping cases that had been brought by the U.S. mills. One stratagem discussed was a Japanese GATT challenge to U.S. antidumping orders, to which the EU would give appropriate backing.

It is unlikely that the framers of the GATT foresaw that their institution could be used to buttress the anticompetitive arrangements of an international steel syndicate. In fact, as the framers feared, the proliferation of clandestine private systems of managed trade will undermine the GATT system just as surely as would the spread of secret bilateral deals between governments. Discriminatory market division arrangements among private firms, like the London Agreement, are antithetical to the letter and the spirit of the GATT itself, including the basic principle of Most Favored Nation treatment upon which the GATT system is based. Having no tariffs but also no trade cannot be regarded as an altogether satisfactory outcome of a half century of tariff reduction efforts--yet this is the prospect that is presented, for example, by the Japan-Canada "non-aggression" arrangement in flat-rolled steel, where, if reports are accurate, there will continue to be no trade between the two countries even if tariffs are completely removed.

Those who expect this case presentation to conclude with a call for the prosecution of those involved in international steel cartels will be disappointed. Criminalizing such conduct will serve no useful purpose and may actually make the problem worse by driving it further underground. The London Agreement and the other cartel arrangements just described are trade distortions, and should be addressed, first and foremost, through the use of trade measures. Over the short term, this means antidumping actions, although such measures, by themselves, are not the whole answer.

Beyond this, OECD governments, both trade and competition authorities, must begin to develop effective ways to address trade problems which arise from private anticompetitive practices. What is impossible to do is to retreat. We cannot be told of cartel activities and merely treat them as a curiosity. Trade liberalization, the GATT, and the World Trade Organization cannot be fully effective without market forces--unfettered by private anticompetitive activities--determining competitive outcomes in the market place.

ALAN WM. WOLFF is Managing Partner of the Washington, D.C. office of Dewey Ballantine. He is co-author of Conflict Among Nations: Trade Policies in the 1990s, as well as other works on trade. He served as U.S. Deputy Trade Representative from 1977-79. This Working Paper is a shortened version of remarks made before the OECD Market Access Roundtable, Paris, June 30, 1994.  

Downloaded from