Producer cartels are about monopolistic coordination aimed at raising the suppliers’ revenues. Efforts to cartelize typically come in waves, and have occurred throughout the history of international commodity trade. This chapter focuses on the 1970s, when the most recent wave occurred. While it lasted, a number of academic efforts were launched to explain the functioning of commodity cartels in general and of OPEC in particular. Despite the theoretical developments and the many modeling exercises that were undertaken, many of the key issues concerning commodity cartelization remain to be fully understood. The remark from an influential survey of commodity cartelization from the mid-1980s (Gately, Reference Gately1984) that “There are a large number of alternative theories, but a much smaller number of sensible applied models” retains its validity even thirty years after it was published.
The present chapter begins by studying the necessary minimum preconditions in terms of elasticities and market shares for successful cartel action. We then identify the markets where these preconditions appear to be fulfilled. There follows an account of the attempts of commodity producers to wield the market power to their own benefit, trying to answer questions such as: What were the triggers for the cartel action? How did it go? How did the buyers react? What prompted cartel disintegration? There is a heavy emphasis on oil, given the extraordinary price performance of this commodity since the early 1970s (Chapter 4), but we express skepticism about the role of OPEC in oil price evolution. The findings of our cartel analysis are not only of interest for the sake of history. They should have a bearing on the future too.
10.1 The Formal Preconditions for Successful Cartel Action
Successful cartelization measures involve either a restriction of supply or a rise in the price charged by the members of the collaborating group, leading to increased revenue for the group. With a given demand schedule, there is a unique relationship between the quantity supplied and the price at which the market is cleared, so the two measures would have equivalent consequences. Where the institutional market arrangements involve producer-set prices, cartel action would ordinarily take the form of an increase in the producer quotations. Where primary commodity prices are set by exchanges, as is increasingly common, the colluding producers could achieve their aim by reducing supply until the desired price level is reached.
Under ideal conditions, producer collaboration should aim at maximizing the joint profits of its members. In terms of Figure 10.1, this would be achieved by reducing supply from Q1, the competitive equilibrium, to Q3, given by the intersection between the collaborating group's marginal cost and marginal revenue. Any output above this level would be unprofitable, because the marginal cost of that output exceeds the marginal revenue. This is the standard profit maximization rule applied by a perfect monopoly. The criterion for successful producer collaboration employed here involves the cruder rule of revenue maximization, which disregards the costs saved by production cuts. Under this criterion, output would be reduced from Q1 to Q2, the latter determined by the marginal revenue of the producer group being equal to zero. Revenue would then rise from P1Q1 to P2Q2. We have adopted this cruder rule for the purpose of the following discourse because we believe that this is about as much as a real-world cartel could aim for. We know no cases of cartels that have defined their supply schedule with sufficient precision, and instituted income transfers between individual members, to make profit maximization a practicable policy.
Figure 10.1 Maximization of profit and maximization of revenue
The present analysis assumes that the participants in the cartel (leading producing or exporting nations in the present analysis) can reach full agreement on a marketing policy that aims at increasing their sales revenue, and that they will adhere strictly to the policy rules. Even these more modest assumptions about the behavioral discipline within the group are somewhat heroic. The possibility of increasing the group's revenue over that reaped in the absence of joint action can be shown to be greater, first, the higher its share in global supply, second, the lower the (absolute) value of the price elasticity of global demand, and third, the lower the value of the price elasticity of outsiders’ supply.
In formal terms, successful market intervention by the cartel requires that the (absolute) value of the price elasticity of demand for its output, EDC, should be less than 1. If EDC is greater than 1, the cartel's revenue will decline when the members jointly raise prices or cut supply The value of EDC is determined by the formula (Radetzki, Reference Radetzki and Helleiner1976):
where
M = the cartel's share of world supply;
EDW = the price elasticity of world demand; and
ESR = the price elasticity of supply outside the cartel.
An EDC which is less than 1 implies that the marginal revenue from the cartel's aggregate supply is negative. Hence, the sales revenue will increase as supply is curtailed. A maximum will be reached when EDC reaches a value of 1 and marginal revenue equals 0. This will happen when supply has been cut to Q2 in Figure 10.1. The smaller the value of EDC, the greater the potential for raising revenue through cartel action.
The success of the collaboration has an important time dimension. This is because the (absolute) price elasticities of world demand and of outsider supply (EDW and ESR) will tend to increase over time, as the final users and independent producers adjust to the conditions caused by the cartel's intervention. The higher prices resulting from cartel action can greatly increase the cartel members’ revenue in year 1 over what they reaped in the competitive market that prevailed in year 0. If the higher price is maintained, their revenue in year 5 may prove substantially lower than in year 0, as a result of the gradual shrinkage of global demand and of the cartel's market share. Present value calculations of the revenue gains and losses over time will be needed to determine the benefit of such a course of events. But a cartel is unlikely to be judged a success unless it manages to keep the members’ revenue above the competitive level for at least several years.
It may be something of a paradox that a cartel which commands no credibility in the outside world will have greater prospects to succeed in its market actions than one which does. This is because if no one believes that the price-raising collaboration will survive, there will be no adjustment to the higher prices resulting from its actions. With no adjustments, the short-run values of EDW and ESR will persist.
The above formula defining EDC can be used to determine the limiting combinations of the price elasticity of world demand, EDW, and the price elasticity of supply outside the cartel, ESR, that have to hold for the price elasticity of demand faced by the cartel, EDC, to be less than 1, and hence for cartel action to increase the collaborating group's revenue. Table 10.1 illustrates the limiting elasticity values for successful collaboration of a group whose share of world supply, M, equals 60%. It will be seen that the potential for revenue-raising action (numbers shown in italics) exists in all cases where the values of EDW and ESR are less than 0.4 (absolute values), but also for selected other elasticity value combinations.
Table 10.1 The price elasticities of demand for output from a cartel (EDC) which controls 60% of world supply (M = 0.6)
| EDW/ESR | 0.1 | 0.2 | 0.4 | 0.7 | 1.0 |
|---|---|---|---|---|---|
| −0.1 | −0.23 | −0.30 | −0.43 | −0.63 | −0.88 |
| −0.2 | −0.40 | −0.47 | −0.60 | −0.80 | −1.00 |
| −0.4 | −0.73 | −0.80 | −0.93 | −1.13 | −1.33 |
| −0.7 | −1.23 | −1.30 | −1.43 | −1.63 | −1.83 |
| −1.0 | −1.73 | −1.80 | −1.93 | −2.13 | −2.33 |
The value of EDC is also related to the range of commodities under the cartel's control, but multi-commodity cartels will gain additional market power (a lower EDC) from their wider coverage only if the commodities are each other's substitutes. No synergies will be obtained from launching a joint cartel by the coffee and copper producers. Since there is little relationship between these two markets, price-raising supply restrictions in one will have no effect on the other. In contrast, the copper producers’ market intervention will be reinforced by a simultaneous restriction in aluminum supply. When copper producers intervene in isolation, the higher copper price will induce substitution in favor of aluminum, and the reduced copper demand will dilute the benefit from intervention. If the copper producers coordinate their action with the aluminum producers so that the prices of both products rise in parallel, no substitution from one to the other will be induced by the price change, and the producers of both metals can reap higher earnings than if each group had acted separately.
The increased market power follows from the fact that the price elasticity of world demand, EDW, is lower for copper-cum-aluminum than for each metal in isolation. The greater the substitutability between the products, the higher will be the benefit of joint cartel action for both, and the lesser will be the possibility to establish a successful cartel for each product in isolation. Attempts to launch a cartel by primary copper producers would be much strengthened if it included a successful effort to get the suppliers of copper scrap on the bandwagon, given the substitutability between primary and secondary metal material.
However, market power is only one of several aspects that determine the prospects for launching successful cartels. Another is the ability to administer and coordinate the members’ actions, so there is a trade-off between the augmented market power and the increased complexity in managing multi-commodity cartels.
10.2 Other Preconditions for Successful Cartel Action
The preceding section clarified why control of a large share of total supply and low price elasticities are necessary preconditions for successful cartel action. That discussion, however, is far from adequate for identifying the commodity markets in which cartelization is feasible.
A first ambiguity arises from the definition of total supply. One could alternatively look at the share of global output under the cartel's control, or at the share of global exports accounted for by its members. The latter figure is usually higher, so it produces a more optimistic impression of the cartel's potential success, but it disregards the possible dilution of the collaborating producers’ market power as independent supply switches between the domestic and export markets.
A second problem is due to the uncertainty and instability of the elasticity values. Price elasticity estimates can vary greatly depending on precisely what is measured, the method used, the time period to which the estimate applies, and the price level at the time of measurement. As noted, long-run price elasticities are usually substantially higher than short-run ones. In terms of Figure 10.1, the demand and supply schedules will be flatter if a longer time period is considered. Price elasticities can also vary with the absolute price levels. For instance, when the demand curve is a straight line, as in Figure 10.1, the price elasticity will rise as prices increase. Hence, measurements of elasticity made at a given time will not necessarily hold if a price change has subsequently occurred.
For these reasons, exercises like that contained in Table 10.1 cannot bring out neat distinctions between commodities that are amenable to successful cartel action and those that are not. Clearly, considerable standard errors are attached to all elasticity estimates, and the best that one could expect from such analyses is a first crude categorization of commodities according to the prospects for monopolistic manipulation.
A third problem concerns the practicalities of producer coordination intended to control commodity supply. Since cartel action is about cuts in supply, the initial issue that needs to be resolved is the overall size of the cut. Well-established producers and producers with above-average cost levels are likely to be interested in greater cuts than new and low-cost producers, who are keen to expand their output. The need to assure full collaboration from producers who jointly account for a large proportion of the total will tend to result in agreements scaled down to suit the convenience of the parties which desire the least proportional output reduction.
Coincidental with the determination of the overall output reduction are decisions about its distribution among participating members. Optimally, only the high-cost output ought to be cut, but to be acceptable such a policy would require income transfers from lower-cost producers, who are allowed to continue their operations, to those who close down. This is rarely if ever practicable, so the sharing of cuts would typically be in proportion to output in the recent past, to actual capacity prior to the proposed action, or to the capacity including expansions in the pipeline. These alternatives usually give rise to protracted quarrels as each cartel participant positions itself to maximize its own yield.
After the joint supply cut has been implemented, each individual member will have a strong temptation to covertly increase its supply and so benefit from the higher price while letting the others carry the burden of restriction. A close inspection of the participating members’ adherence to the agreement will therefore be needed to prevent it from breaking apart.
A few important inferences for the practicability of international commodity cartels can be drawn from the above. The smaller the group of participating producing or exporting countries needed to attain the required share of world supply, the simpler will it be to reach and maintain a supply restricting action. Agreement will be much easier to reach and administer in a group of four or five than in a group of 12 or more. Similarity among the participants will also facilitate monopolistic coordination. If they are of equal size, have matching cost structures and levels, pursue similar goals, and operate in comparable social and political environments, an agreement will be easier to reach than when there are great differences within the group. The ease with which output can be cut and supply can be monitored will also affect cartel operation. The cohesion and trust within the collaborating group will benefit from transparency of the burden sharing.
Empirical studies of international cartel endeavors in commodity markets have often been simplified by regarding countries instead of producing corporations as participants. Individual producers are not always easy to identify, and the volume of their exports may be hard to quantify. Identification and quantification are much easier to handle at the national level. There are also some more fundamental arguments in favor of treating national governments instead of corporations as cartel members. First, in many countries corporations are subject to anti-trust legislation, which makes their overt participation in cartel action difficult. Governments are not subject to such restrictions. Second, governments have and often do exercise sovereign powers to regulate exports as they deem fit. A third motivation for viewing governments rather that corporations as the key cartel players is that the 1970s, the decade in which there were strong and widespread beliefs in “producer power,” and numerous commodity cartels were launched, was just preceded by or coincidental with a great wave of nationalizations of resource industries, predominantly, but not exclusively, in developing countries (see Chapter 11). Efforts to intervene in commodity markets were often initiated by governments, with the newly nationalized corporations used as instruments for policy implementation.
The question of whether governments or producing corporations are the more efficient executors of cartel policy in international commodity markets has been discussed for decades, but remains unsettled. Summarizing experiences from the inter-war period, Rowe (Reference Rowe1965) concludes that an effective international commodity control scheme could be secured only with the active participation of governments. While the empirical evidence from petroleum, bauxite, phosphates, and uranium in the 1970s (see below) supports Rowe's conclusion, opposing views have been aired. For instance, Grilli and Yang (Reference Grilli and Yang1988) assert that effective collusion is easier to achieve by a group of private profit-maximizing agents that can act in a covert manner than for governments with a variety of national goals, whose actions by necessity become a “semipublic international political affair.” The dynamic of cartel launch sometimes follows a path where leading private corporations initiate the process and then approach producing country governments to act as cartel fronts. This was the case in 1974, when Rio Tinto Zinc initiated discussions with the governments of Chile and Zambia (among others), about a copper production cut, with the subsequent action handled by CIPEC, the Intergovernmental Council of Copper Exporting Countries.Footnote 1 A similar course of events took place somewhat later in the uranium market, where the leading private producers coordinated their marketing efforts, using the Canadian government as their visible front (Radetzki, Reference Radetzki1981).
The characteristics of commodity markets that are amenable to successful price-raising actions by producers can now be summarized, and the potential candidate commodities which meet the required criteria picked out. The method used will be that of successive elimination.
Reasonable prospects for cartelization require a low (absolute) price elasticity of demand. The commodities must not be easily replaceable by close substitutes. This excludes the group of edible oils and their raw materials, which are easily interchangeable, and whose production is so dispersed that a joint product cartel would hardly be feasible. The same is true for fruits like bananas, apples, and oranges.
Another precondition for successful cartel action is that the price elasticity of outside supply should be low, at least over a perspective of 3–5 years. This would exclude quite a number of commodities, e.g., the cereals group and sugar, whose production could be speedily expanded in many places in response to higher prices that looked like persevering for couple of years. The same is true for products like cotton, jute, and possibly wool.
After these eliminations we are left with rubber, the tropical beverages and most minerals, all characterized by limited substitutability and extended periods required to create new production capacity. One would now like to fine-tune the price elasticities of these commodities to get a better grasp of the prospects for market control, but given the sizable standard errors that surround existing measures of elasticity, we deem such an effort to be futile.
The level of supply concentration might throw at least some additional light on the issue under scrutiny. With all else equal, a high level of concentration among producers (whether nations or corporations) should facilitate supply coordination. Tables 2.5 and 2.6 in Chapter 2 contain data from which information on concentration in global export flows can be distilled. Only some of the products listed pass the arbitrary criterion we have chosen: that the five most important countries should account for no less than 60% of global exports (in Table 10.1 above, M = 0.6 is used to distinguish commodities that might satisfy the preconditions for viable cartel endeavors). The products and their shares work out as follows: Cotton 77%; Coal 81%; Iron ore 82%; Rubber 73%; Rice 73%; Tin 62%; Wheat 67%; and Wool 71%. We have used the same source as that employed for constructing Table 2.5 (http://comtrade.un.org/) to add some further products with heavily concentrated exports in the early 2010s, and find Cocoa 62%; Maize 88%; Nickel 75%; and Tea 60% to pass the test. The cartel prospects for cotton, rice, maize wheat, and wool are unfavorable on account of their high short-run supply elasticities (see above), leaving a group of seven among the commodities investigated with some prospects for producer market intervention criterion. We note with some surprise the low concentration for three products with cartel action history: Copper 51%; Oil 42%; and Coffee 36%.
Cartel prospects hinge not only on national supply concentration, but also on corporate concentration. Even though most competition authorities prohibit collaboration across corporations to raise prices, cartels still frequently occur in both national and international markets. Table 10.2 lists commodities where the five largest producers control more than 60% of global supply. Intriguingly, they are all found in the mining industry.
Table 10.2 Five leading corporations’ share of global production
| Year | Share of global tonnage, % | Comments | |
|---|---|---|---|
| Beryllium | 2012 | 90 | Largest company controls 87% |
| Diamond | 2013 | 69 | Based on value |
| Iron ore | 2013 | 70 | Based on seaborne trade |
| Niobium | 2013 | 99 | |
| Platinum | 2013 | 80 | |
| Zirconium | 2012 | 63 |
Further insights into the role of concentration for cartel action would, however, require more profound analyses of the affinity of the producers, the structure of the export market, and the industrial organization of the buyers of each commodity. Successful cartel action would be less likely where the buyers are few, financially powerful, and able to retaliate.
10.3 Actual Experiences of Commodity Cartels in the 1970s
The popularity of commodity cartels appears to occur in waves, usually triggered by one or several outside events, but the cartels that attempt to establish monopolistic prices are seldom long-lived, and tend to disintegrate as a consequence of stagnant demand and rising independent supply, both prompted by aggressive price policy.
In the 1930s, a number of price-raising international commodity cartels were established by producers in agricultural as well as in mineral commodity markets, somewhat counter-intuitively in response to the exceedingly low price levels that reigned during the Great Depression (Rowe, Reference Rowe1965). The monopolistic actions were widely viewed with sympathy and were overtly supported by the governments of the consuming countries, including the US government. Higher prices were seen as essential for the maintenance and expansion of commodity production, sometimes even for the survival of producers, and, at a wider level, for the restoration of world prosperity (Herfindahl, Reference Herfindahl1959). These cartel efforts were overtaken by events following the outbreak of the Second World War, with ensuing scarcities and far-reaching government controls.
Another wave of commodity cartel action occurred during the 1970s, this time in response to the combination of widespread nationalizations of mineral resource industries, following third world independence from colonial bonds and a very strong boost in commodity demand in 1972–74, triggered by the global macroeconomic boom in those years. A widespread perception of commodity power emerged among producers, especially in the developing world, and efforts were launched to establish producer associations, predominantly in the minerals field, with price raising as the primary goal. The most important and persistent was the oil cartel. The spectacular and lasting price increases in oil reviewed in Chapter 4 have been commonly attributed to OPEC's market interventions, and so aroused a lot of enthusiasm among other commodity produces. Producer efforts to raise prices in non-oil markets were successful in some cases, though short lived, and the failure was often due to shrinking demand for the cartel's output, as the longer run price elasticity proved to be disappointingly high. In other cases, no visible price impact can be detected from the attempts at market intervention.
Bauxite
In the late 1960s, Jamaica began to urge the governments of bauxite-producing countries to form an association for the exchange of information, reduction of rivalries, the establishment of a joint front to the multinational aluminum companies, and coordinated increases of export taxes (Brown, Reference Brown1980). Enthused by the apparently successful collaboration within OPEC, but also by the booming demand for their product, the bauxite-producing countries founded the International Bauxite Association (IBA) early in 1974. By 1975, its members accounted for 85% of nonsocialist world (NSW) output. The production units were still largely owned by the vertically integrated aluminum companies, and there were not really any meaningful market quotations for the product. The cartel, therefore, largely operated through the increase of production and export taxes.
Jamaica's government was also the first to take action. At the time, the country was the world's second largest producer, and, on account of transport distances, it enjoyed a considerable cost advantage in the US market. In 1974 and 1975 the government instituted a very sharp increase in its production levies and export taxes that went far beyond its locational monopoly. As a result, the import cost in constant money of Jamaican bauxite in the USA roughly doubled between 1973 and 1976 (Vedavalli, Reference Vedavalli1977), and continued to increase until 1980 (World Bank, 1994).
The Jamaican government apparently expected that the other members of the IBA would follow suit, so eliminating the relative loss of Jamaica's competitiveness. To some extent, this also occurred. Surinam instituted fiscal levies similar to Jamaica's. Guinea, too, raised its bauxite taxation, but by less than the two Caribbean countries. However, Australia, the world's largest producer, and an IBA member, refused to join in these interventions.
Table 10.3 reveals an apparent depletion in that the Caribbean producers’ competitiveness over time resulted in a substantial loss of their market share. The main gainers were Australia and Guinea, members of the IBA who were more concerned about their sales, and Brazil, which never joined the association.
Table 10.3 Bauxite output among leading producers in the nonsocialist world (NSW)
| 1974 | 1982 | 1990 | |
|---|---|---|---|
| NSW total (m tons) | 71.3 | 66.7 | 99.9 |
| Jamaica (%) | 22 | 12 | 11 |
| Surinam (%) | 10 | 5 | 3 |
| Guinea (%) | 11 | 18 | 16 |
| Australia (%) | 28 | 35 | 41 |
| Brazil (%) | 1 | 6 | 10 |
The falling market shares of Jamaica and Surinam would have been easier to handle in an expanding market. In fact, the NSW demand for bauxite fell by 6.5% between 1974 and 1982, mainly due to the extended recession in the wake of the 1973–74 oil crisis. This speeded up the erosion of the cartel.
The frequent alterations of the Jamaican taxes and levies in the 1970s and 1980s, along with other concurrent changes implemented in its bauxite/alumina industry (e.g., production controls, nationalizations) make it difficult to isolate the impact on the government's revenue from the bauxite levies. Nevertheless, the price-raising interventions must be deemed a failure. The country's share of the NSW market declined from 22% in 1974 to 12% in 1982, with no subsequent recovery. The Caribbean policies clearly favored Australia and Brazil, which declined participation in the market management. Jamaica, the original founder of the IBA, formally withdrew its membership in 1994, and the association collapsed soon after (Crowson, Reference Crowson and Maxell2006).
In terms of the formal analysis in Section 10.1, the cartel's lack of success was caused by EDC being too high in the medium term. The very low value of EDW was overwhelmed by a low M (the initial market share of the Caribbean producers) and a high ESR. The advantage of the Caribbean nations’ resource endowment was not pronounced enough to give them durable market power.
As an afterthought and to reflect on the remarkable long-run flexibility of (even exhaustible) raw materials supply, consider for a moment the evolution of events from the early 1990s, when the IBA collapsed, until the present. In 1990, NSW bauxite output amounted to 100 mt, and the four IBA members listed in Table 10.1 still supplied 71% of that total. By 2013, the market had expanded to 230 mt, but the share of the four had shrunk to 34%, while formerly insignificant suppliers had grown to dominate the market. In that year, Indonesia supplied 24% of the total, China 19%, and India almost 9% (WBMS, 2014).
Phosphate Rock
Booming demand and the apparent success of OPEC led to a decision by the state-owned Moroccan phosphate rock producer Office Chérifien des Phosphates to raise its producer price, from $14 to $42 per ton in January 1974, and then again to $63 in July of that year (UNCTAD, 2000). In the short run, this intervention was highly effective, because the state-owned phosphate enterprises of Algeria, Togo, and Tunisia, the mixed-ownership producer in Senegal, and the members of the US export cartelFootnote 2 Phosrock raised their list prices in close concert with the Moroccan action. The entire group accounted for more than 70% of global phosphate rock exports at the time, almost half of which was from Morocco (UNCTAD, 1981).
The price-raising scheme proved short-lived. In 1974 itself, exports from all the participants in the scheme increased significantly, so the price gain was exacerbated by gains in volume. In 1975, however, a severe world recession reduced demand. The higher prices also resulted in deferred farmer demand and substitution in favor of other fertilizer raw materials. EDW proved to be quite high, and the cartel was unable to withstand the strains that emerged, despite its high market share. The Moroccan phosphate rock price was reduced to $49 in 1976 and $38 in 1977 (UNCTAD, 2000). In constant dollar terms, the 1977 price was at par with levels in pre-cartel days.
UraniumFootnote 3
The international uranium mining industry entered the 1970s in a state of profound depression. It had been built to satisfy the huge military demand during the 1960s. With military needs fully satisfied by the end of the decade, the existing uranium capacity was far in excess of nuclear reactor needs for many years into the future. The low prices did not provide full cost coverage for a large segment of the industry, so many producers left the business.
The depressed market was the trigger that brought producers together in an effort to safeguard their survival. A series of meetings initiated by the government of Canada took place in 1971. The governments of France and South Africa were represented, and leading private producing companies from a number of countries took part. The meetings were intended to “put some order into the international uranium market … to coordinate uranium production and marketing policies” (Nucleonics Week, 1971).
This embryo of the uranium cartel was quite frail while the market remained weak. The most it could do was to reduce rivalry among members, and to issue directives aimed at preventing further price falls. At the end of 1973, however, a number of unrelated but coincidental factors completely reversed the market situation. The most important of these was a decision by the US uranium enrichment agency (at the time a state-owned virtual world monopoly) to change the rules under which it marketed its services. According to the new rules, enrichment had to be commissioned decades in advance of actual needs, and there were high penalties for cancellation. Owners of existing and planned nuclear reactors signed up to excessive enrichment contracts, and then went on a buying spree to secure their future uranium requirements.
Having institutionalized their collaboration in the preceding years, the uranium producers responded by temporarily withdrawing from the market, so the prices exploded. The spot quotation went up from less than $7/lb U3O8 in late 1973, to more than $40 by mid-1976, in spite of an NSW output increase of 15% between the two years. Prices in long-run contracts signed in this period followed suit. The producers reentered the market only after prices had reached the $40 level. The cartel worked under very favorable conditions. The private producers were actively supported by the governments of the major exporting countries. The group accounted for a high proportion of the NSW supply, but the precise level of M is hard to establish given the covert nature of much of the cartel's operations. It faced a price elasticity of demand (EDW) that was close to zero. New capacity to produce uranium would take a long time to establish, and in the meantime ESR remained quite low. So the prices stayed very high through most of the 1970s.
The subsequent decline was caused by an increasing realization among the nuclear utilities that they had greatly overcommitted themselves to uranium purchases, given the shrinking plans to expand nuclear capacity. Demand for newly mined uranium was sharply reduced as the excessive inventories held by the nuclear power generators were scaled down. New production came on stream by the end of the decade, and discoveries of large and very rich uranium deposits in Canada and Australia altered earlier perceptions of impending scarcity. After five years of exceedingly high profitability for the industry, the prices in constant money were back to the levels that had prevailed before the cartel burst to life.
Copper and Iron Ore
Two further attempts at establishing commodity cartels in metal mineral markets need to be mentioned, but they can be treated quite briefly, since they both failed to institute effective price-raising measures (Crowson, Reference Crowson and Maxell2006).
CIPEC, the Intergovernmental Council of Copper Exporting Countries, was formed in 1967 by the governments of Chile, Peru, Zaire (later renamed Congo-Kinshasa), and Zambia for the purpose of raising prices through collective interventions in the copper market. Yugoslavia and Indonesia joined later, while Australia and Papua New Guinea became associates. Like the bauxite and phosphate exporters, the governments of the main copper-exporting countries were enthused by OPEC's apparent success, and CIPEC tried in 1974–76 to raise prices with the help of production cuts, but the efforts failed due to mistrust among members, and because the eight members controlled too small a share (37% in 1975) of global mine supply. Additionally, there was a large-scale supply of copper scrap over which the primary producers had no control. CIPEC subsequently dwindled in importance with the collapse of production in Zaire and Zambia, and the withdrawal of several members. It was formally dissolved in 1988, and its then remaining functions were taken over by an International Copper Study Group, formed in 1993.
The Association of Iron Ore Exporting Countries (APEF) attempted in 1975 to set export prices. The effort was unsuccessful, first, because two important members, Australia and Sweden, were unwilling to go along, and second, because Brazil and Canada, both sizable export suppliers, refused even to join. APEF reduced its role to collecting statistics on market trends until its demise in 1989.
In 2014–15, a few years after the end of the commodity boom, and in the face of a large increase in iron ore production accompanied by dwindling iron ore prices, the voices of a possible iron ore cartel were raised once more. The currently fourth largest iron ore producer, Fortescue, recently argued for a “managed cartel,” where the export of Australian iron ore producers would be “predetermined” (Hurst, Reference Hurst2015). This would be an attempt to control supply from the Australian export market. As indicated in Table 10.2, the conditions for forming a cartel in the iron ore market are favorable, as the producer concentration is relatively high. The three largest producers, Rio Tinto, BHP Billiton, and Vale, currently control 58.3% of seaborne traded iron ore. Furthermore, these three companies have almost equal shares of the market. However, despite these conditions, there is so far little evidence that a cartel will be formed. Our doubt rests on the fact that the three largest producers are also the players with the lowest production costs; thus it is not they, but the smaller producers, that suffer most as prices decline.
OPEC
The Organization of Petroleum Exporting Countries (OPEC) was brought into existence in 1960.Footnote 4 Its major purpose was to form a united front in an attempt to arrest the fall in the member governments’ revenue per barrel (Griffin and Steele, Reference Griffin and Steele1986). Until the early 1970s, the organization led a dormant life in a market situation characterized by excess supply.
By the early 1970s, the market had strengthened substantially due to the very rapid growth of world oil consumption (8.3% compound annual growth between 1960 and 1972).Footnote 5 The world macroeconomic boom of 1972–74, during which all primary commodity prices exploded, permitted the OPEC governments to find acceptance among the multinational oil firms that exploited their oil to very large posted price rises, hugely increasing their fiscal revenues, while the oil companies passed the increase on to the final consumers. With the very low short-run price elasticity of demand for oil, there was little need for downward supply adjustments in response to the higher price. Given the speedy demand growth, one might well argue that the market would have accepted the price rises and maintained them in the short- to medium-term even in the absence of OPEC and without coordination among the oil multinationals.
The instrument used by the OPEC group during most of its life to restrain output and strengthen prices has been a formal production quota system, instituted in 1983 and maintained ever since. Quotas could certainly strengthen prices, at least while they lasted. However, besides typically having a short duration, the quota system was highly disjointed. The price goals and the timing of quota application have shifted in an erratic manner. A variety of rules governed the establishment of quotas for each member – e.g., historical output, production capacity, reserves, production costs, or population – and these rules changed over time. Variously motivated exceptions from quotas have also been common, where individual members were permitted to produce without restrictions. Additionally, the shallow nature of the production restraints and constant cheating by most OPEC members made a significant price impact of OPEC's actions unlikely. A comparison of actual OPEC output with the quota ceilings between 1983 and 2001 (Molchanov, Reference Molchanov2003) reveals that production exceeded the ceiling by 6.9% on average, with numerous occasions when the excess ran up to 15% or more. Production in excess of permitted quotas continued after 2001 (Laherrere, Reference Laherrere2011). Ironically, full compliance has been achieved only during episodes (like 2005–06) when the production ceiling itself tested the limits of each member's available production capacity, such that cheating by excess output was not feasible. In fact, data over the past decades (IEA, monthly) reveal that, excepting Saudi Arabia (the dominant OPEC producer), Kuwait, and UAE, virtually full technical capacity utilization has been the rule among the group's members.
With about one-third of OPEC output and a production policy often independent of the rest of the cartel, Saudi Arabia's actions have on several occasions had a strong price impact, but the objectives have varied over time in a somewhat incoherent manner (Aguilera and Radetzki, Reference Aguilera and Radetzki2016). One instance occurred early in 1979, when the Iranian revolution had pushed up prices strongly, and the Saudis implemented a sizable (though short-run) output reduction, resulting in a further price explosion to levels in real terms not seen since the 1860s. Another took place in late 1985, after a five-year period during which the Saudis virtually on their own defended an excessive price that severely depleted their market share. A sizable output increase led to a price collapse early in 1986, and an eventual recovery in the demand for the country's oil. A third instance relates to the Saudis’ attempt to arrest the price rise following from the production lost due to the war between Iraq and Kuwait in the early 1990s and to the simultaneous Russian oil industry malaise after the USSR breakdown. Between 1989 and 1992, the Saudis increased their output from 5.6 to 9.1 MBD, thereby greatly reducing the price boost from Iraqi, Kuwaiti, and Russian production declines. It should be added that between 1988 and 2010, Saudi production persistently exceeded the quota it had agreed to (Laherrere, Reference Laherrere2011). Yes, prices have been influenced, sometimes strongly, by Saudi Arabia's interventions, but not uniformly upwards, and certainly not by the cartel's collective action.
For a credible long-run impact on prices, the market interventions of a colluding producer group would have to include constraints on investment and capacity expansion. Such measures, however, have never been applied collectively by OPEC. As discussed in Chapter 4, two circumstances with no relationship to the cartel's market management, have in fact brought about a general constraint on capacity growth in OPEC members, with some even experiencing sizable capacity shrinkage.
The nationalization of large parts of the oil industry in most OPEC countries in the course of the 1970s resulted in a very severe loss of efficiency and an enduring inability to undertake complex investments by the young state-owned firms (Radetzki, Reference Radetzki1985). Furthermore, the ensuing greedy extraction of most of the profits earned by these firms for the benefit of government budgets left inadequate funding for investments (Aguilera and Radetzki, Reference Aguilera and Radetzki2016). This added to the difficulties in raising production. The inability to expand capacity had a far more potent influence on market prices than the cartel's production quotas.
Table 10.4 depicts the lagging performance of OPEC as a whole in the world oil industry. The group's share of global output in 1973, a not very impressive 51%, declined to only 42% 40 years later. Its contribution of less than a quarter to global output growth is remarkably low, given that almost three-quarters of global reserves are located in OPEC countries, and that they comprise the huge and economically exceptional ones of the Middle East.
Table 10.4 OPEC and world oil output (MBD)
| 1973 | 2013 | Change | |
|---|---|---|---|
| OPEC | 29.9 | 36.8 | +6.9 |
| World | 58.5 | 86.8 | +28.3 |
| OPEC share (%) | 51 | 42 | ?24 |
A more focused and even more dramatic evolution in production capacity due to the debilitating effects of the resource curse in selected OPEC countries is presented in Table 4.3 and the surrounding text of Chapter 4, to which the reader is directed.
We assert that state ownership, government greed, and the resource curse in combination adequately explain the exceptional performance of oil prices since the early 1970s, and that OPEC's collective actions have not yielded any significant lasting price impact.
Is our vision of OPEC's impotence credible, given the predominant popular view of the oil cartel as outstandingly successful, even forming a precedent for other commodity groups? As a matter of fact, a number of serious studies, undertaken at different points in time, appear to share our vision. Thus, MacAvoy (Reference MacAvoy1982) argued that the observed trend of oil prices could be adequately explained by a competitive model. In the same vein, Alhajji and Huettner (Reference Alhajji and Huettner2000) contended that statistical tests fail to support a cartel model of OPEC behavior, while Smith (Reference Smith2005) found the “evidence” of OPEC behaving as a price-raising cartel inconclusive. The study by Bina and Vo (Reference Bina and Vo2007) is more assertive and assures that OPEC is neither a cartel nor exhibits any sign of market domination, market control, or monopoly. Finally, adopting a 50-year perspective, Gately (Reference Gately2011) concludes that ever since the early 1970s, OPEC has not exploited its market power to raise prices, although its capacity stagnation has certainly had long-term effects on world oil prices. OPEC's limited ambitions at market and price management could be the explanation for the group's survival for such a long time.
10.4 Conclusions
Not many commodity markets are amenable to successful monopolistic collusion. The necessary but not always sufficient formal conditions are low price elasticities of demand and of outside supply. Even when these conditions are fulfilled, concentration among suppliers and a considerable degree of cohesion is essential for success.
The urge to launch cartels comes in waves, and different circumstances can trigger their establishment. During the 1930s, depressed prices prompted cartel action by bringing together producers that faced a survival threat. Third world independence in the 1960s and the subsequent nationalization of resource industries established a firm belief in “producer power” that resulted in concerted monopolistic interventions in many commodity markets in the 1970s. Experiences from these periods reveal that price-raising cartels normally have a short life. Government participation appears essential for launching and maintaining price raising intervention. Even then, cartels tend to disintegrate after some years, as the critical elasticity values increase over time.
OPEC is exceptional in that it came into being and still exists at an age of over 50 years. But then, we have argued (convincingly, we hope) that its survival is due to its impotence as market mover, and that the exceptional performance of oil prices is due to circumstances other than cartel action.
