The Common Market Organization (CMO) for cotton was introduced in 1981 with Greece's accession to the EU. Its purpose was to support cotton production in those regions where it was an important part of agriculture and to offer a fair farm income. The system protected cotton growers from the major price fluctuations seen on the world market while allowing processors (i.e., ginning firms) to dispose of EU cotton fibre at world market prices. There are no restrictive measures on trade with third countries and access to EU market is free. On the other hand, there are no support measures for exports, such as refunds.
The CMO for cotton is organized on the basis of three policy measures, i.e., deficiency payment, corresponsibility levy, and maximum quantity guaranteed (MQG), and involves both cotton growers and cotton ginners. A guide (target) price is set by the Council and its difference from the world price determines the level of production subsidy (aid per tonne) that producers receive every year. Since unginned cotton is a non-storable, non-tradable good, its world price is determined through the ginned cotton price on the basis of yields from ginning, which vary between 30-40 percent in terms of the fibre obtained after processing. However, the amount of aid per tonne is not paid directly to cotton growers. It is actually paid to cotton ginners on the condition that farmers have received a minimum price per tonne of unginned cotton. This minimum price is set slightly below the target price and refers to standard quality unginned cotton defined on the basis of its impurity and moisture contacts, and in terms of the length and grade of fibres. For the period 1995-1996 to 2002-2003, the world price accounts for one third of the minimum price paid to producers and the production aid for the rest two thirds.
The minimum price is associated with the corresponsibility levy mechanism and the MQG. That is, application of the guide and minimum prices set by the Council was limited to a MQG that the Council also set. If actual production exceeds the MQG, a stabilizer mechanism reduces the minimum price, as well as the guide price, by the amount of the corresponsibility levy and hence it also reduces the amount of aid per tonne received by cotton growers. The level of the corresponsibility levy is related to the amount actual production exceeds the MQG. In this way the budgetary cost of aid may be controlled within certain limits, as long as there are no significant reductions in the world price of unginned cotton.
Even though the CMO for cotton was affected neither by the 1992 CAP reform nor by the changes in support scheme for producers under the Agenda 2000, it went through several changes mainly with respect to the corresponsibility levy and the MQG mechanisms. In particular, the MQG was set at an EU level during the period 1981-1995 while it has been set at a national level for each Member State since then. Similarly, the level of the corresponsibility levy was determined per fixed amounts of excess production during the period 1981-1995 but has been measured as a percentage base of excess production since then. Nevertheless, the main principles of the CMO have remained unchanged.
In the period prior to Spain and Portugal accession to the EU (1981-1985), the MQG was set at 560 000 tonnes and the corresponsibility levy, when activated, it could result in a 1 percent decrease of the guide price for every 15 000 tonnes by which the MQG is exceeded. During this period, when Greece was the only country concerned by the CMO of cotton, production gradually rose from 33 000 to 52 000 tonnes. However, the annually set MQG was never exceeded and thus the scheme operated as a pure deficiency payment.
During the period 1986-1991, the first change in the CMO of cotton took place as the MQG was increased to 752 000 tonnes because Spain was also a major cotton producer. Even though the corresponsibility levy mechanism remained unchanged, a cut-off point below which the guide price could not be reduced was introduced in order to protect cotton growers from the risk of very big falls in the minimum price from one season to the next. Community production during this period (ranging from 875 000 to 1 200 000 tonnes) always exceeded the MQG, activating the corresponsibility levy mechanism.
Another change in cotton regime occurred during the period 1992-1995. In particular, the way of determining the corresponsibility levy changed. The 15 000-tonne tranches for establishing the percentage reduction of the guide price were replaced by a coefficient calculated using the overrun on the MQG. This new formula was fairer for production levels that were near the 15 000-tonne changeover point. In the period 1992 to 1995, Community's production continued to exceed the MQG, reaching 1 335 000 tonnes. However, due to drought, a uniform reduction of the guide price throughout the EU was viewed as unjust by Spanish growers. Political pressure on this put forward the changes that took place next.
During the period 1995-2000, the MQG increased to 1 031 000 tonnes and it was divided into national quantities guaranteed (NQG) to make growers in each Member State more accountable for their production. The NQG for Greece was set at 782 000 tonnes and for Spain at 249 000 tonnes. These NQGs were calculated on the basis of the average production of each Member State in the last consecutive three-year period in which weather conditions were considered normal. However, to guarantee budget neutrality, the guide and the minimum prices were reduced proportionally. In addition, the corresponsibility levy mechanism was revised again. Under the new scheme, the price reduction was 0.5 percent for each percentage point by which a Member State's production exceeded its NQG. In any case, the overall support for the sector was expected to be equal to or above €770 million.
During the period 1995-2000, Greek production exceeded its NQG by 54 percent on average and Spanish production exceeded its NQG by 30 percent on average. This means that the corresponsibility levy mechanism was activated during the whole period. Application of the scheme resulted in minimum prices reported in Table 2, which are well below the guide (target) price of €1063/t and the official minimum price of €1010/t. The difference in the minimum prices paid to cotton growers between Greece and Spain were the outcome of their respective levels of production. In Spain, the minimum price paid to producers was more stable and higher than in Greece because Spanish production was more stable and exceeded its NQG by a lesser amount.
In addition, during the period 1995-2000, the cut-off mechanism was activated to ensure that overall support would be not exceed €770 million. This mechanism was implemented three times in the case of Greece (i.e., crop years of 1995-1996, 1997-1997, and 2000-2001) and twice in the case of Spain (i.e., crop years of 1997-1998 and 2000-2001). If the price reduction cut-off had not been in force, expenditures for the aforementioned crop years would have been €638, €701 and €632 million, respectively compared to the actual cost of aid reported in the last column of Table 2. Thus, the average reduction of the guide (target) price during the 1995-2000 period was 22 percent in Greece and 15 percent in Spain, which were smaller than if the price reduction cut-off had not been in order. Nevertheless, Community's production continued to rise reaching a record of 1 760 000 tonnes.
Since 2001 the corresponsibility levy mechanism has been strengthened. In particular, above a Community production threshold of 1 500 000 tonnes, divided between Greece (1 138 000 tonnes) and Spain (362 000 tonnes), the 0.5 percent price reduction for each percentage point overrun of the NQG was increased by a further 0.02 point for each extra 15 170 tonnes grown in Greece and each extra 4 830 tonnes grown in Spain.
The unginned cotton market is depicted in Figure 1. Given that unginned cotton is a non-storable, not directly consumable good, which can only be used as a raw material for ginned cotton, it is reasonable to assume that the demand for unginned cotton is perfectly elastic at the world price level (Karagiannis, Katranidis and Velentzas, 1997). On the other hand, S represents the supply curve of unginned cotton and QM is the MQG or the NQG. Given the aforementioned discussion of the EU cotton policy regime, the kinked demand curve PTaD corresponds to the effective demand for EU producers (Herruzo, 1992), where PT is the guide (target) price. Eventually producers receive the target price PT as long as national production does not exceed QM. In the case that actual production exceeds the NQG, the mechanism of the corresponsibility levy is activated and the resulting minimum price is determined by the intersection of the downward sloping portion of the kinked demand curve and the supply curve. In the case depicted in Figure 1, cotton growers receive price P as the minimum price because their production Q exceeds QM. The difference between PT and P is the rate of the corresponsibility levy. The kinked demand approaches asymptotically the world price PW for large amounts of actual production indicating that at the limit the guide price may coincide with the world price.
For the period prior to 1986 the supply of Greek cotton growers, who were the only Community producers then, was to the right of S, actual production was smaller than the MQG and the cotton regime was operated as a deficiency payment. As the gross margin for cotton was larger during this period, a large portion of cultivated land was diverted to cotton production from other arable crops. This gradually shifted the supply curve to the right. In addition, Spain's accession to the EU had also contributed to this rightward shift of the aggregate supply curve. Cotton growers then had two options: either to comply with policy regulations and receive the entire guide price, or to violate the MQG and receive a lower price.
Acting as price takers, individual cotton producers perceived the demand for their produce to be perfectly elastic and arranged their individual production according to the guide price, not the aggregate MQG or the NQG (Karagiannis and Pantzios, 2002). This may be due to the competitive nature of cotton sector and asymmetric information between farmers, EU authorities, and national governments. As a result production increased beyond and producers were receiving a price lower than the guide price. This behaviour of individual producers confirms an assertion made by Burrell (1987) that, unless imposed on an individual basis, restrictions on the aggregate production might not be effective. The division of the MQG into the NQG in 1995 may be seen as a partial reaction to this assertion.
More importantly, Karagiannis and Pantzios (2002) have shown that (for the case of Greece) cotton growers could have been better off if they had complied with the NQG. That is, compliance with (rather than consistent violation of) the NQG could be an absolute Pareto improvement, in the sense that farmers would have gained a higher per capita income, in terms of producers' surplus, and taxpayers' cost would have also been lower. This can be shown with reference to Figure 1. Producers' surplus, if farmers had complied with policy regulations, would have been PTadkO but it actually was PckO. Karagiannis and Pantzios (2002) estimated the difference - bcd + PTabP to be positive in the case of Greece for the cropping years 1995-1996 to 1997-1998. On the other hand, taxpayers' cost would have been PTahpW if farmers had complied and PcgPW otherwise. Again they estimated the difference PTabP - bcgh to be positive. As a result, deadweight loss would have been lower if farmers have complied with the NQG and thus policy implementations could have been more efficient.
