position with millions of small sellers and a handful of big buyers. The market for coffee, for instance, is dominated by a few big buyers like Nescafe and Volcafe, while there are millions of producers and sellers of coffee.
     The problem of many small producers is compounded by the entry of new producers into the market. For instance, in the market-based reform in Vietnam, large numbers of farmers took to coffee production as a cash crop. These new producers may be willing to accept a lower net income, one which is an improvement for them but lower than what tra­ditional producers are used to. In the manner that sections of manufacturing have relocated to areas of lower wage costs, sections of commodity production can also relocate to areas where small producers are willing to settle for less than what traditional producers earned. This type of com­petition among small producers in developing countries is a feature of current commodity production.
     While the above analysis puts competition among pro­ducers as a key factor in low and fluctuating prices for com­modities, another analysis points to the non-cognizance of environmental costs as leading to low prices for commodi­ties (Dasgupta and Maler, 1990). When the social costs of production are higher than private costs, there is a subsidy on the basis of non-valuation of environmental resources, which are production resources for the small producers.
     More recently, yet another factor, that of competition between developed and developing countries, has entered the picture in leading to low prices of some commodities. This is the case of those commodities, like cotton and sugar, which can be produced in both tropical and in temperate or semi-temperate conditions. Take the case of sugar, which can be produced both from tropical cane and temperate beet. In the EU sugar producers are paid twice what they would get in the international market. At the same time, EU also export about 5 million tonnes of sugar at the lower world price, thus "undermining further the price received by farmers in developing countries" (Robbins, 2003). But there are still many commodities, like coffee, tea or ba­nanas, which are not grown in OECD countries and are thus not affected by subsidies and protection by OECD countries.
     While commodities are largely produced in a competi­tive environment, the markets for manufactures are much more monopolistic, leading to a secular decline in terms of trade for commodities (Singer, 1950). Productivity increases are passed on to consumes and buyers as prices fall. But in the case of manufactures, the monopolistic market position of producers enables them to keep the benefits of technolog­ical advances that lower costs of production. This analysis leads to the conclusion that the only development path for developing countries is to diversify away from commodity production into manufacture; something that many Asian countries have successfully accomplished.
     Along with providing some adequate return to labor in commodity production, there is also the problem of en­abling transitions in commodity production. If incomes are assured then will there be a shift from high-cost to low-cost producers? This is the positive function of the competitive market mechanism. The market by itself, however, brings about this transition in an entirely ruthless manner, lead­ing to the destitution of the displaced producers. To bring

about an effective transition commodity interventions have to meet two objectives: providing a reasonable income to the producers and enabling an orderly transition from high cost to lower cost producers.
     There are two types of interventions in commodity mar­kets: (1) price stabilization, with buffer stocks; and (2) out­put regulation, with quotas and restrictions. The second, however, requires a few producers or a few organized groups of producers, to organize a cartel. This is what OPEC has done successfully to control crude oil production. Prices are then allowed to take their own levels.
     In the Bretton Woods Conference, where the IMF and World Bank were born, Keynes proposed a commodity board which would operate buffer stocks. At each point of time, a base price would be set and fluctuations allowed of 10% on either side of the base price. If, however, stocks in­creased at the end of the year, then the base price would be marked down by 5% and vice versa for decreases in stocks. The annual reduction of prices would have the effect of en­abling lower cost producers to increase their share of pro­duction, while higher cost producers would leave the sector. But unlike the violent adjustments of the market system this adjustment would be brought about in a gradual and thus less painful manner. There could be stability, but not stagna­tion, as producers enter or leave the market.
     The US was opposed to such a scheme that would stabi­lize commodity prices and it was not taken up after Bretton Woods. But some commodity boards did come up in the post-Second War period. Among these commodity boards only the coffee board had what is known as an economic clause, meaning it would undertake market stabilization activities. The others confined themselves to trying to set export quotas.
     The major problem with export quotas is that they tend to freeze production among existing producing. What about new countries that wish to enter? A big factor in the collapse of the International Coffee Agreement (ICA) was the spread of production outside the countries with quotas. Indigenous peoples in upland areas (e.g., Vietnam or India) have often taken to coffee production as a substitute for forms of swid-den, or introduced coffee into the swidden mix. All this de­veloped production centers outside of the traditional coffee growing areas. It is not possible to manage these changes in the location of production through export quotas.
     In the 1970s there was a series of UNCTAD-inspired commodity agreements. But the experience of these com­modity boards has not been encouraging. In price deter­mination they played a positive role, in that coffee buyers purchased coffee from a handful of boards rather than large numbers of small operators. But the quotas were used, particularly in the African countries, to favor those eth­nic groups from which the ruling sections came (Gibbon and Ponte, 2005). In years where high prices were paid for products, the amount paid to farmers was kept low in order to increase the amount retained by the boards. In years of low prices the price stabilization efforts were swamped by exchange rate fluctuations. Large resource transfers were needed in those years, well beyond the capacity of the gov­ernments concerned.
     The commodity agreements were discontinued because the main consuming countries withdrew financial support.