Was it a credible commitment by De Beers to threaten to punish any defectors from the diamond cartel

Cartel agreements are “an institutional form of collusion”1. They present an alternative to traditional models of oligopolistic interaction, such as Betrand or Cournot models, that assume that firms act solely in order to maximise their total profits. In cartels, firms cooperate in order to maximise their joint profits, eventually leading to higher profits from each individual firm involved. More specifically, firms agree to produce smaller quantities, in order to drive up the market price.

Although this sounds ideal in principle, in practice there exists the incentive for firms to deviate from such agreements and produce more than their predetermined quota in order to reap the benefits of a high price on a larger scale. This will place downwards pressure on the artificially high price. Ideally, firms must be able to retaliate in the case of deviation in order to force firms to maintain their agreed quota. However, such a potential threat must be backed with credibility: it must be worthwhile for firms to punish defectors, otherwise the threat will lose its strength and deviation will be inevitable.

The validity of threats can be examined using the same framework, but from the point of view of the firm enforcing the cartel agreement. Assuming that deviation and punishment will be followed by a reinstatement of the previous cartel agreement and that no punishment will lead to a disintegration of the cartel agreement, it is again possible to use game theory to analyse the firm’s strategic options: “punish” or “do not punish”. If the firm has a low discount factor it will attach little value to future cash flows, and hence will not have incentive to punish.

On the other hand, the threat to punish will be greater with a higher discount factor. Therefore, when determining the likelihood of retaliatory action, a firm will have to be aware of the enforcer’s discount factor, and determine the most probable strategic choice in the case of deviation. In the case of De Beers, it is possible to analyse the factors affecting the discount factor. The probability that the industry will exist in the next period is high, driving up the discount factor.

Officially, prices are only set once a year, which would indicate a lower ‘f’. However, De Beers has the capacity to make sudden changes to prices by flooding the market with extra diamonds, a factor which will artificially raise f. As for growth in demand, it can be interpreted as being relatively stable, given that diamonds have remained an expensive commodity as a result of De Beers’ propaganda that “diamonds are forever”. The interest rate varies worldwide in the markets that De Beers is involved in, and hence can be ignored for the sake of simplicity.

Overall, this points towards a relatively high discount rate. This indicates that De Beers will probably be willing to carry out retaliatory action against deviant firms in order to ensure that it reaps its expected future profits. An additional point worth mentioning is that De Beers is not only willing to carry out threats, but it is also capable of doing so: it possesses a stockpile of diamonds, as well as the financial resources to buy back excess supply on the market.

Because firms in the diamond industry tend not to be involved in other market areas, their exposure to such potential retaliatory action can be very high, especially because in many cases De Beers is more powerful than the firm or country in question. The credibility of De Beers’ threat to punish defectors can be observed not only in theory, but also in practice. In the case of Zaire, it clearly decided that the short term losses resulting from a flooding of the market were worth the prospect of making Zaire join the cartel again.

The same can be observed by the fall in prices it caused for Australia’s Argyle. The Russian case, on the other hand, is slightly different: given the sheers size of the Russian diamond production, De Beers is in a more difficult situation. However, it has continued to attempt to gain control, by buying controlling shares in diamond fields and relying on Russia’s need for hard currency.

The only exception to De Beers’ rule of retaliation is Angola, which was spared because it “was never perceived as a long-term threat to the CSO but rather as a product of the political turmoil in country at that time. 3 Because it did not have a great impact on future cash flows due to the limited time span of the situation, De Beers did not believe it to worth punishing. Ultimately, De Beers’ approach is consistent with the game theory applied to multiple interactions between firms, using the discount factor to determine the value of future cash flows, and proves the validity of its threat to punish firms and countries that dare to defy its power.