Special report No 8/93 concerning the common organization of the market in raw tobacco together with the Commission's replies

Special report No 8/93 concerning the common organization of the market in raw tobacco together with the Commission's replies Official Journal C 065 , 02/03/1994 P. 0001 - 0072 SPECIAL REPORT No 8/93 concerning the Common Organisation of the Market in raw tobacco together with the Commission's replies (94/C 65/01) (Observations pursuant to the second subparagraph of Article 188c (4) of the EC Treaty)>TABLE> 1. INTRODUCTION 1.1. The Community support for tobacco growing and first processing was introduced by Council Regulation (EEC) No 727/70 (). In the first year of application Community expenditure amounted to some 5 Mio UA () and increased to about 362 Mio ECU in 1981.

In 1982 Community expenditure doubled and followed a steeply increasing trend until in 1991 it reached its historical high of 1 329,6 Mio ECU. In 1992 expenditure fell back slightly to 1 232,9 Mio ECU but increased again in 1993 with a budget of 1 274 Mio ECU. This amounts on average to 3,5 Mio ECU of Community expenditure per day, with average support of about 6 000 ECU for each hectare. On this basis tobacco is by far the Community's most expensive crop. 1.2. With smoking coming under increasing public criticism because of its disastrous implications for public health and health-related social costs, this level of financial support looks out of place, especially when the Community itself finances a 'Europe against cancer` programme albeit with a modest annual budget of 1,5 Mio ECU for anti-smoking campaigns.

Since 1993, regulations require a contribution from producers, deducted from the premium they receive, to finance research and information into the dangers of tobacco use. It is obvious that the massive support of tobacco production within the Common Agricultural Policy contradicts both the Community and international health policy and initiatives. For, while the Community imports some 70 % of its manufacturing requirements of baled tobacco, it exports about half this quantity, which represents some 50 % of the Community's high cost production of tobacco. 1.3. This report examines the evolution of Community spending on tobacco and its effect on tobacco production in the Community since the Court's last report on this matter in 1987 ().

It also compares the Community system to the United States of America one which only involves some 30 Mio USD for administering a tobacco scheme which produces twice the Community's annual output. 1.4. The audit was carried out at the Commission and in the main producing Member States, Germany, Greece, Spain, France and Italy, at the central and regional administrations and at the premises of processors and growers. It also used information provided by third parties such as the World Health Organisation, the tobacco industry, the International Tobacco Grower's Association and the specialized press.

The Court wishes to thank the United States Department of Agriculture and the General Accounting Office for its excellent collaboration. 1.5. The Court regrets that the Commission was not as forthcoming as it ought to have been in response to repeated requests from the Court for a copy of a report on an investigation ordered by the Commission into allegations of irregularities in the Commission's tobacco division. The Commission still maintains that investigations in the matter, which began in 1988, are not yet completed. An auditor of the Court was belatedly permitted by the Commission in April 1993 to read (but not to copy) a report after the Court's audit fieldwork had been finalised and that report contained information giving rise to suspicions of fraud.

On the basis of that examination there was no evidence of action on the part of the Commission or a Member State to bring suspected wrongdoers to justice. 1.6. The Court has requested the Commission for a copy of a contract with a company, connected with a retired Commission official, which conducted that investigation and for details of the payments made thereunder but such information has not been given to the Court. 1.7. The Court also requested the Commission to furnish a copy of a report of an external expert on the classification of Italian Geudertheimer tobacco but the Commission, despite reminders, has refused to give a copy of the report to the Court.

1.8. To overcome, to some extent anyway, the obstacles created by the Commission's lack of cooperation, the Court has been obliged to expend additional manpower and time to obtain required information elsewhere. Because of the matters already known to the Court and to avoid any further delay, it is publishing this report at this stage. The Commission should make full disclosure to the Court of all material information and take, in cooperation with concerned Member States, all necessary remedial action. 1.9. The report contains observations on the management and control of the tobacco market which is carried out by both the Member States and the Commission's own services.

Many of the Court's findings have considerable financial consequences. The Court therefore urges the Commission to take all necessary steps to seek reimbursements where due and to improve controls and general management without delay. 2. THE MARKET FOR RAW TOBACCO The product 2.1. Tobacco originates from the sub-tropical regions of Central and South America. It exists in many varieties (the French Tobacco Research Institute in Bergerac watches over the genetic heritage of a thousand varieties and hybrids) and is adaptable to quite different climatic and soil conditions. 2.2. Tobacco seeds are sown in seed beds during late winter, with the small plants being transplanted to the fields in Spring.

Tobacco growing is labour intensive, requiring careful soil preparation, weeding and for some varieties intensive irrigation. Quality is also dependent on the density of plants in the field. Furthermore, tobacco should ideally be grown in rotation, with periods of up to seven years between two cultivations on the same soil in order to avoid tobacco diseases. 2.3. Harvesting is spread over several weeks from the beginning to the end of summer. Some varieties (e.g. Virginia) are picked by leaf stage according to the ripening process, whilst others (e.g. Burley) are harvested by cutting the whole plant.

Methods of harvesting tobacco also vary from region to region. 2.4. After harvesting, the leaf tobacco is cured by the growers before it is sold to the processors. The European Common Market Organisation (CMO) which covers only raw tobacco and not the end products involves 34 varieties which are classified by curing method into five groups: (a)flue cured varieties are dried in ovens or barns where temperature, circulation and humidity of air are controlled via flues; the best known variety in this group is Virginia, the largest single type used in American blend cigarettes; because of their golden yellow colour they are also called Bright varieties;

(b)Burley and Maryland, also used in American blends are light air cured in barns and not fermented (exception: the Spanish fermented Burley); such varieties are also called blond or brown tobacco; (c)Badischer Geudertheimer, Havanna and Paraguay are the best known dark air cured varieties; curing is similar to the light air curing but the tobacco is allowed to ferment up to several months; such black tobaccos are used in cigars and dark cigarettes; (d)sun cured varieties not only include the commercially demanded Greek oriental varieties Basmas and Katerini but also the low demand tobaccos, Tsebelia and Mavra which are also grown in Greece;

(e)Kentucky (grown in Italy) and Nijkerk (grown in France) are the best known representatives of the fire cured varieties; they are dried in the smoke of open (oak) fires and are used in blends and for snuff. 2.5. After curing the tobacco is sold to about 300 first processing undertakings in the Community. First processing means any kind of treatment to turn the leaf tobacco into a stable conditioned product ready for storage and manufacturing. It entails the removal of tobacco debris and impurities, the sorting and mixing to leaf position, colour, texture and quality and baling. It may also include fermentation for some varieties and, if the tobacco manufacturer requires it, stripping, which removes the stalk and larger veins from the leaf.

The world market 2.6. According to the latest figures available at the Commission (see Table 1) the world production of raw tobacco amounts to 7,4 Mio tonnes, of which the Community produces 430 300 tonnes (about 5,8 %). The world's biggest tobacco producer is the Peoples' Republic of China with some 39 % of world production, followed by the United States of America with about 10 %. 2.7. Only about 21 % of world tobacco production is traded on the world market, with the Community's share being 37 % of total imports and 15 % of total exports. The world wide demand for raw and processed tobacco is dominated by four large multinational tobacco conglomerates from the United States and the United Kingdom ().

This situation is, however, not obvious, as these companies act via a multitude of subsidiaries as buyers of leaf and processed tobacco all over the world. 2.8. In contrast to the domination of the market by the very limited number of buyers, the supply is characterized by a huge number of individual producers. In the EEC there are about 180 000 tobacco growers, whilst the USA has over 130 000 farm families directly engaged in tobacco cultivation. In developing countries, whose share of the world tobacco production has gone up from 53 % in the early 1960s to 75 % in the late 1980s, tobacco cultivation generates large-scale rural employment.

China, has some 16 million growers and in Zimbabwe the tobacco industry is the largest employer of labour. Worldwide about 33 million people are engaged in tobacco growing (). 2.9. At the end of the 1980s world tobacco production was broadly in balance with supply. In 1990 consumption fell slightly short of production and is not forecast to overtake it until 1995. Indeed, according to the 'Economist Intelligence Unit` (), world production of leaf tobacco is forecast to rise on average by 1,1 % per year while consumption will rise by 1,4 %. However, to the extent that manufacturers continue to reduce their stocks as a proportion of their consumption, there will also be a slight increase in stocks held by processors.

2.10. Although about 61 % of all tobacco products in the world are still produced by state monopolies, these will lose importance, as the recent opening of markets in Turkey, the states of the CIS and the Far East by the multinational tobacco conglomerates shows (). The increasing internationalism of the Anglo-American tobacco industry and the subsequent advance of American type blends will increase the proportion of flue-cured tobacco (now 59 %) and light air-cured tobacco (now 11 %) in world production notwithstanding their comparatively high growing and curing costs. 2.11. Despite a decline in consumption in Western and Eastern Europe, the states of the CIS, Japan, North America and Australia, increasing total population in the rest of the world combined with the increase in urban population in developing countries will, on a worldwide basis, reverse the trend noted in the developed countries.

The EC-market >TABLE> 2.12. In the Community, tobacco provides income for about 180 000 farmers cultivating approximately 200 000 hectares of this crop. Table 2 shows production by group of tobacco and Member State in 1992. The largest producer is now Greece, followed by Italy and Spain. They together account for 90 % of Community production. 2.13. Before the creation in 1970 of the CMO for the raw tobacco market, demand was balanced in Spain, Greece, France and Italy by the then existing State tobacco monopolies and the relative closure of the tobacco markets in these Member States. When the CMO for raw tobacco was introduced the monopolies over growing and processing but not over manufacture had to be dissolved () and the markets opened, not only for tobacco from other Member States but also from third countries, where prices are lower because of low wages.

As transport of unprocessed tobacco over long distances is quite unusual because of its instability, world trade is nearly exclusively in baled tobacco. Customs duties on baled tobacco are consolidated in the General Agreement on Tariffs and Trade (GATT), so that tobacco enters the Community with only low duties or, according to the Lome convention, with no duties at all when coming from ACP-countries. Therefore the CMO provided Community producers with guarantees for their income equivalent to those afforded under the previous national marketing arrangements. 2.14. Producer interests in the EEC have been increasingly channelled through cooperatives which often also own first processing facilities.

This, together with the safeguard of the CMO, limits the impact of the tobacco manufacturers' power on the growers' incomes. The Common Market Organisation (CMO) 2.15. The CMO for raw tobacco as introduced in 1970 worked for about twelve years without major problems and changes, with the exception of measures introduced in 1978/79 aimed at controlling supplies. In 1981 the accession of Greece increased the Community's total tobacco production by some 40 % and introduced oriental varieties which had not been grown before in the Community of 9 Member States and for some of which a real market did not exist.

The next enlargement of the Community to Spain and Portugal in 1986 increased the Community's production by a further 14 %. Both enlargements did not modify the CMO principles, but increased the Communities production of varieties for which demand was shrinking. >TABLE> 2.16. The CMO goals include maintaining farmers' income, reducing unwanted surpluses by adapting production to varieties in demand and keeping production and population in less favoured farming areas. These goals became more and more difficult to reconcile. In order to maintain the high level of institutional prices (norm prices and intervention prices) set by the Council each year in a market without outside protection, the CMO had to protect the tobacco producers' outlets in the Community against cheaper tobacco imports.

At the same time the CMO also had to secure the manufacturers' necessary supplies of full flavour varieties from the world market. 2.17. It is for this reason that the CMO was modelled on a deficiency payment system, whereby a premium was supposed to 'secure the norm price and ensure the sale of the tobacco produced in the Community`. The rate should be set at a level taking into account past and foreseeable outlets on the Community market as well as the influence of trends in prices for tobacco imported from third countries. Thus, to a certain extent the premium had to cover the difference between the lower world market prices of varieties in competition with home grown varieties and the higher EC norm prices.

It should allow the first processor (or buyer) to pay the farmer at least the intervention price in the form of a fixed but quality related minimum price and to sell tobacco to the manufacturers at prices comparable to those prevailing on the world market. 2.18. Accompanying measures had to ensure that surpluses without an immediate market outlet could either be exported at world market price level or temporarily stored. So the CMO foresaw export refunds and intervention storage. In 1988 rapidly increasing budgetary expenditure led to the introduction of stabilizer measures in the form of maximum guaranteed quantities (MGQs) associated with limited reductions in premiums and prices triggered when these quantities were exceeded.

2.19. Finally, on the 14 October 1991 the Commission presented to the Council a proposal for the reform of the CMO for raw tobacco which the Council adopted on 30 June 1992 (). The reformed CMO still involves 34 varieties grouped in five classes according to the curing method and with three further classes for the Greek sun cured varieties Basmas, Katerini and Kaba Koulak (classic). The principle innovations in the new CMO are: (a)norm prices are abolished and price setting is left to the market; (b)premiums are no longer fixed for each variety, but are fixed on the basis of one for each group of varieties depending on the method of curing e.g. sun cured, flue cured etc.;

(c)premiums continue to be paid to the processor, but have now to be passed in full to the grower; (d)premiums are paid for tobacco meeting certain minimum quality characteristics, if the moisture content differs from the level fixed for each variety, premiums are paid on the adjusted weight; (e)the maximum guaranteed quantity of 390 000 tonnes is replaced by a maximum global guarantee threshold of 370 000 tonnes for the 1993 harvest and 350 000 tonnes for the following years; (f)this global threshold is split by Member States into quotas which are distributed to first processors or to producers; no premium is paid for quantities which exceed the individual quotas;

(g)with the aim of concentrating supply and adapting it to the qualitative requirements of the market premiums can be increased by 10 % for production under contracts with producer organisations; (h)intervention storage and export refunds are abolished; (i)control agencies are to be established by the Member States; (j)to facilitate the creation of inter-branch organisations () in the tobacco sector, in order that the various operators involved in production, processing and marketing can contribute, inter alia, to changes in economic behaviour and market balance; (k)a conversion programme for four problem varieties;

(l)a research programme for varieties with lower tar content. 2.20. The following chapter records the results of the audit of the accounts and underlying transactions until 1992 which were still based on the CMO as it was before the reform. As outlined in Council Regulation (EEC) No 2075/92, transitional measures could be applied in 1993 obliging Member States inter alia to allocate processing quotas to processors and to reduce in two steps the Maximum Global Guarantee Threshold to 350 000 tonnes. The reformed CMO will therefore not be completely effective until 1994 and expenditure on export refunds and intervention storage will still have to be disbursed up to 1997 on tobacco harvested before 1993.

Provision is also made for the premium and production limitation system to run until 1997 and then to be reviewed and possibly adapted in the light of experience. 3. EXAMINATION OF THE ACCOUNTS AND OF THE UNDERLYING TRANSACTIONS IN THE MEMBER STATES 3.1. Table 3 shows the budgetary expenditure incurred by measure between 1987 and 1992. In general the accounts from which these figures are drawn reflect expenditure managed and declared by Member States. Table 4 shows expenditure incurred by both measure and Member State. 3.2. There are certain features of the tobacco CMO which affect the way in which the expenditure is presented in the Commission's consolidated accounts i.e.:

(a)premiums paid in a particular year relate to harvests of that year and prior years; in its forecast on expenditure on tobacco () for the year ending 31 December 1993, the Commission estimated that 1 111,3 Mio ECU out of a total of 1 268 Mio ECU would concern premiums on the harvests of the two preceding years; >TABLE> (b)the payment of a premium may only take place once the leaf tobacco has been put under supervision, processed and sold to manufacturers or been exported, this may take several years; advances of the premium may be paid to processors during this period; however, advances of premiums paid in a particular year are presented as definitive expenditure although the tobacco concerned had not left supervision; for instance, in 1992 the Greek authorities made final payments of premiums for tobacco that had been placed under supervision in Greece between 1981 and 1986; >TABLE> (c)Commission Regulation (EEC) No 2822/88 () introduced immediate depreciation at the time of purchase at a rate of 65 % of the buying-in value after deduction of the premium; another 'special` depreciation is operated at the end of the financial year i.e. in October; this depreciation is based on estimates of stocks supplied by the Member States in August and takes into account appropriations available at the year end; the high level of depreciation in the years prior to 1991 led to substantial profits on sales of stocks in 1991 and 1992;

(d)expenditure shown year by year is not definitive until the clearance procedures for a particular year are closed, i.e. three years later; Table 5 contains the disallowances in favour of the Community budget decided by the Commission following the clearance of the accounts of 1987 to 1990; (e)profits on sales of intervention stocks are set off in the accounts as negative expenditure, thus understating total expenditure on tobacco for a particular year; disallowances of expenditure following clearance audits are also set off as negative expenditure from total EAGGF expenditure thus understating EAGGF expenditure; both methods lead to an automatic re-employment of funds of previous financial years for the same purposes.

3.3. Following the Community accounting rules (), Member States split the amount paid for purchasing intervention tobacco into a notional premium part and an intervention part. The premium part is declared in the monthly declarations of expenditure as premium and is booked by the Commission as if it was a real premium paid for tobacco sold on the market. The difference between the notional premium payment and the purchase price of the tobacco constitutes the stock value at purchase. It does not disclose to the budget authority the true cost of tobacco not finding a market outlet. >TABLE> 3.4. As in other markets, the aid (the premium) is fixed in ECU but paid in national currency by the Member States.

Under the tobacco CMO, processors were eligible for an advance of premium when tobacco was taken under supervision, but the premium did not become definitively payable until the date that the tobacco left supervision after processing. This led to unjustified expenditure. Member States with a national currency that had depreciated between the date of the advance and the date that the premium became definitively payable would make a further payment in national currency which represented the difference between the premium converted to national currency at the original rate of exchange and the premium converted at the new rate of exchange.

3.5. Under the reformed CMO the problem has been resolved. The premium becomes due when the tobacco is delivered by the grower for processing, and the date for applying the conversion rate is fixed. However, the problem persists for harvests prior to 1993. The estimated budgetary cost of this additional payment has been some 50 Mio ECU per annum. The cost will continue for the next four years until the original CMO finally ends. 3.6. There is no financial justification for the practice. The Commission should have proposed changes to the original CMO to avoid this additional expenditure. Unless the regulation is amended there is likely to be an effect on the smooth operation of the reformed tobacco CMO.

Processors will be encouraged to retain tobacco from harvests prior to 1993 in order to benefit from any devaluation. Furthermore, unjustified budgetary expenditure will be incurred. Examples of the effects of devaluation in particular Member States are given in paragraphs 3.34 - 3.36 and 3.84 - 3.85. Premiums 3.7. Premiums are the most important category of Community expenditure for raw tobacco, representing about 90 % of all expenditure under the CMO. They are paid by the responsible national authorities to first buyers who either process or on whose behalf tobacco is processed under the condition that they have paid at least the intervention price (minimum price) to the grower.

Premiums are paid on the quantity of raw tobacco put under official supervision at the processor's premises or remote collection points, at a standard moisture content and after the deduction of sub-standard tobacco and extraneous material. 3.8. As the ultimate right to the premium payment depends on processing the tobacco delivered, the final amount of the premium can only be determined when the processed tobacco 'leaves supervision`. This can be several years later but since 1991 the time allowed has been limited to 4 years (). Therefore storage of leaf and processed tobacco, together with receipts into and disposals out of stock, at the processors premises has to be supervised by the competent national authorities.

3.9. In order to enable the processor to pay growers the minimum price at the time that he buys the tobacco, the Community regulations foresee advance payments of the premium. They are subject to the existence of a 'European cultivation contract` (which is a purchase/sales contract) between the processor (first buyer) and the grower and to the lodging of a guarantee of 20 % of the amount of the premium by the processor. The grower is required to sell the entire production of the areas indicated in the contract. 3.10. Since 1990, Member States have also had the option of making the premium payment conditional on the existence of cultivation contracts between grower and first processor (buyer).

Since 1991 this option has become obligatory (). Under Article 2b of Commission Regulation (EEC) No 1726/70 (1), contracts between growers and processors, as well as cultivation declarations by growers who process their own tobacco, usually via a grower cooperative, are required to be registered with the national authority responsible for control. Under the terms of Article 2c of the same Regulation, the national authorities have to carry out on-the-spot checks of 5 % of the areas grown with tobacco (), in order to verify the information in the cultivation contracts. 3.11. Every year, the Council Regulation fixing prices sets the production zones in which the cultivation of each variety is permitted.

In addition, since 1988, under Article 7 a(1) of the basic Regulation () prices and premiums are only applicable for varieties which originate in a commune where they have been cultivated at least once in the last five years. Furthermore, the commune must be indicated in the cultivation contracts. From 1992, as a quality related measure the premium payment was limited to quantities corresponding to yields fixed by variety and zone (). 3.12. When tobacco is taken under official supervision by the competent national authority, a moisture test should be carried out. Commission Regulation (EEC) No 409/76 () requires the selection from each bale of a number of leaves proportional to its weight, and that the number of leaves be sufficient to be properly representative of the bale as a whole.

The sample should include equal quantities of leaf from the outside of the bale, leaf from the centre and leaf from an intermediate position. The leaves selected must be mixed together and chopped up (cutting width of between 0,4 and 2 mm). The chopped leaves should then be mixed and a representative sample must be taken. The moisture content should be established according to either the Brabender or the Beaudesson scientific method. If such laboratory tests are not used, 5 % of the advance premium payments have to be withheld by the paying agency until the final payment is established. In order to avoid overpayment on tobacco that is too wet, or underpayment on tobacco that is too dry, an adjustment of the weight has to be made if the measured moisture differs by more than 1 % from the reference moisture content laid down in Annex IV of Commission Regulation (EEC) No 1727/70 ().

Furthermore the portion of impurities and substandard tobacco must be determined and deducted. 3.13. The payment of premiums depends on the presentation by the processor to the paying agency of a formal proof that the tobacco has been taken under supervision. This proof consists of a premium certificate established by the control authorities in the Member States. The first part of this certificate is completed when tobacco is taken under supervision and the second part when the tobacco leaves supervision. 3.14. Although some checks (e.g. moisture checks) are prescribed in great detail, not all of the control procedures necessary to ensure correct payments are laid down by Community regulations.

Member States, therefore, have adopted a variety of control methods to suit their own purposes. 3.15. Article 8 of Council Regulation (EEC) No 729/70 () on the financing of the Common Agricultural Policy stipulates that: 'The Member States... shall take the measures necessary to: - satisfy themselves that transactions financed by the Fund are actually carried out and are executed correctly; - prevent and deal with irregularities; - recover sums lost as a result of irregularities or negligence.`Thus the national control systems for tobacco must contain a certain number of key-controls, whether or not they are fixed in specific Community regulations.

The essential controls should consist of checking: (a)the veracity of data contained in the cultivation contracts/declarations, e.g. area cultivated and yields; (b)the eligibility of production zones and communes from which tobacco is taken under supervision; (c)tobacco taken under supervision against details in the cultivation contracts; (d)the variety and the quality of tobacco delivered; (e)that the minimum price has been paid; (f)that the security of stocks of raw and processed tobacco under supervision is assured; (g)the processing and disposal of tobacco; (h)the correctness of payment orders.

3.16. During processing the bales of raw tobacco supplied by farmers lose their identity, due to sorting, blending, stripping, etc. As the final premium payment is conditional on proof that the processed tobacco has left supervision, the national control systems have also to guarantee that the tobacco leaving supervision was obtained from tobacco taken under supervision in a given year. This is done by a comparison of the varieties, qualities and weight of tobacco leaving supervision with those of the tobacco taken under supervision. Therefore, tobacco leaving supervision has to be checked in the same way as tobacco coming under supervision.

The Court believes that such a comparison is impossible. 3.17. Finally, Member States are required, under Council Regulation (EEC) No 4045/89 (), to carry out post payment checks, called scrutinies, which are based on the commercial documents of companies receiving more than 200 000 ECU per year from the EAGGF guarantee fund. Germany 3.18. In Germany the Community tobacco scheme is controlled by the Federal Ministry of Finance via its customs directorate whose principal customs office in Hamburg (HZA-Hamburg Jonas) is the paying agency under the terms of Article 4 of Regulation (EEC) No 729/70 ().

Twenty-one processors prepare the tobacco for later sale to manufacturers. The tobacco processed is mostly grown in Germany although some is imported from other Member States, principally Italy. Tobacco is taken under supervision at remote collection points by local customs officers. Customs staff also control the processors. 3.19. The registration of cultivation contracts is the responsibility of HZA Hamburg Jonas. However, the audit revealed that contracts were not registered, but simply filed by year and processor, and that premium certificates did not refer to contracts, thereby making it impossible to relate the certificates, on which advances had been paid, to their corresponding contracts.

The following anomalies were observed: (a)an advance premium of 2,6 Mio DM had been paid to the German buyer of Italian tobacco based on 15 premium certificates which did not contain any reference to a final contract and which the German paying agency could not, in any case, produce at the time of the audit (June 1991); at the processor's premises, where the tobacco had been fermented on behalf of the buyer, customs documents relating to these premium certificates were found which referred to the date of a contract, but no supplier was indicated; a contract was found however, which indicated that the lot had been supplied by an Italian grower cooperative; but at the Italian paying agency 'AIMA` the same contract reference number referred to other tobacco which had been contracted for by the same German processor with a different processor/dealer in Italy; the Italian investigating authorities were informed of the discrepancy by the auditors and the German authorities accepted a formal request from the Italian authorities to carry out an audit at the German buyer's premises; the German authorities informed the Court in July 1993 that no irregularities were identified in Germany; the Italian authorities should continue to follow up the irregularities discovered to a satisfactory conclusion;

(b)on a copy of another contract between the same buyer and an Italian dealer which is under investigation by the Guardia di Finanza, the tobacco variety was amended by hand from 'Forchheimer Havanna` to 'Badischer Geudertheimer`; the original of this contract examined at AIMA did not contain this amendment; however, as the Council had decreased the premium for Italian Badischer Geudertheimer to the level of Forchheimer Havanna the contract amendment had no financial consequences; (c)three contracts were noted which failed to meet the requirements of Article 3(1) of Regulation (EEC) No 727/70 () which states that premium is payable to persons or companies who buy their tobacco directly from growers or their cooperatives and not via an intermediate dealer; the three contracts in question indicated that the tobacco was not directly purchased from the growers by the German processors; therefore no premium should have been paid.

3.20. The German authorities check the eligibility of communes from which tobacco is delivered for taking under supervision only as part of the post-payment checks under Regulation (EEC) No 4045/89 () and, under the terms of Article 2(c) of Regulation (EEC) No 1726/70 (), when the authorities carry out on-the-spot checks of the 5 % of the areas grown with tobacco (). France and Italy have introduced computer programmes which carry out checks on the eligibility of communes prior to payment, thereby enabling them to detect non-eligible supplies of tobacco before payment and thus avoid recovery procedures.

Community funds are better safeguarded under such a system. 3.21. Checks when tobacco is taken under supervision are carried out by representatives of both the grower and buyer under the supervision of local customs officers. The moisture check is carried out by feeling and examining the tobacco leaves according to a procedure called 'haptische Methode`. Any tobacco which is considered to be too moist is rejected for further drying by the grower, all other tobacco is accepted and recorded as having the standard moisture content laid down by the regulations. The German authorities consider their moisture checks to be accurate within approximately 2 % but tests performed by the Court's auditors in the processor's own laboratory revealed that the tobacco just taken under supervision at 'standard moisture content` was between 3,7 and 7,6 % drier, resulting in an underpayment of the minimum price to the grower.

As the payment of the minimum price is a stringent condition for receiving the premium, checks on moisture content should be improved and the payment of minimum prices should be checked before payment of the premium. 3.22. HZA-Hamburg Jonas carried out checks on premium certificates to ensure that minimum prices were paid. However, the checks were meaningless in respect of the variety Badischer Geudertheimer, which accounts for about one third of the German tobacco crop, because the qualities and consequently the prices indicated in the premium certificates could not be relied upon. All deliveries of this variety are recorded on the premium certificates as being class III quality, which has a lower price than class II.

However, during visits by the Court's auditors to collection centres and processors' premises it was discovered that some of the tobacco was in fact class II. Therefore, for this tobacco there is no assurance that the minimum price had been paid. 3.23. There is indeed a general problem with verifying the minimum prices paid by processors for tobacco grown in Germany. The check is not carried out at the time the tobacco is taken under supervision but during the scrutiny audit under Regulation (EEC) No 4045/89 () one to two years after final payment and then not systematically. No check at all was made for small beneficiaries and, for tobacco imported from Italy, there was no confirmation that the minimum price was paid to the Italian growers.

The German authorities had been notified of the problem by the Commission, the last time in 1990, yet the German system had not been modified at the time of the Court's audit, in November 1991, so as to ensure that the payment of the minimum price was checked prior to granting the premium. 3.24. According to Article 7(2) of Commission Regulation (EEC) No 1726/70 () the first claim from a processor for a payment of premium must be accompanied by a stock declaration. The German authorities only check these declarations when the processors are selected for scrutiny under Regulation (EEC) No 4045/89 (1).

This is clearly unsatisfactory as a physical check of the stocks declared is no longer possible at that time. Greece 3.25. In Greece the tobacco scheme is managed by the Greek tobacco board (EOK), a public body which authorised the cultivation of tobacco, before the accession of Greece to the Community. It has its head office in Athens and has various regional offices in the tobacco growing areas. It inspects growers' crops from sowing and planting to harvesting. It also controls tobacco processing, which takes place in 122 processing plants, nearly all of them situated in the north of the country.

EOK employs a large number of agronomists and inspectors in order to perform its service tasks and to meet its control functions. 3.26. EOK also checks claims and authorizes payment of the premium. The payments themselves are made by the paying agency DIDAGEP, a directorate of the Ministry of Agriculture. 3.27. EOK is a self-financing body. Part of its income is provided by a 2 % levy on the value of purchases of leaf tobacco imposed each year on processors (). The total amount of levy received during the period 1987 to 1991 exceeded 56 Mio ECU (7 700 Mio DR) and represented 38 % of EOK's total income.

3.28. The problems of carrying out checks on areas cultivated are exacerbated by the fact that no land registers exist in the tobacco growing regions in Greece. Therefore, it is not possible in contracts to separately identify each plot of land by using cadastral references. 3.29. The Greek authorities do not check that premiums are only granted for varieties of tobacco from eligible communes. This has led to particular problems for the production of Virginia in the regions of Agrinion and Thessaloniki. In 1989, when Agrinion accounted for 2 875 tonnes, (30 % of total Greek Virginia production of 9 480 tonnes), 13 of the 29 communes growing Virginia were not eligible for premiums.

In 1990 the figure rose to 24 out of 54 communes. In Thessaloniki 8 out of the 25 communes receiving premiums in 1989 and 8 out of the 32 communes in 1990, were not eligible for premium. Yet, in both cases premiums were paid without checks being made. 3.30. It is evident that premium payments have been incorrectly disbursed on tobacco coming from non-eligible communes. The Greek authorities did not carry out the checks laid down in the regulations for all varieties planted since 1989. Notwithstanding being reminded by the Court of their obligation, the authorities failed to carry out the necessary checks.

However, the Court estimates that the quantity involved may be about 1 000 tonnes of Virginia in both 1989 and 1990 and smaller quantities in the following years (as communes become automatically eligible in the year following the first harvest of a new variety). As there are likely to be similar irregular payments in respect of other tobacco varieties, the total amount incorrectly paid is likely to be substantial - probably in excess of 4 Mio ECU. 3.31. The premium certificate is not established according to the rules. Its first part is only established at the end of the harvest, and not, as required by the regulation, when tobacco is actually taken under control.

Instead 'sub-premium certificates` are prepared for lots of tobacco delivered from the various village collection centres to the processor. EOK officials admitted that tobacco may even be processed before it was included in a 'sub-premium certificate`. 3.32. In such circumstances, no reliance can be placed on the data contained in the final premium certificate, as checks on leaf tobacco cannot be carried out after it has been processed and proper supervision cannot be exercised if tobacco is processed before it has been entered into the premium certificate. EOK has undertaken to improve procedures.

3.33. Control of first processors and exporters is based on records kept by inspectors in official registers which are outside the traders' bookkeeping systems. There is no check between these and the books of account and the financial records of the undertakings to ensure both the authenticity and the accuracy of the transactions entered in the registers. In the absence of such reconciliation the veracity of the data on which payments are based cannot be confirmed. 3.34. Although the procedures for checking tobacco leaving supervision were generally satisfactory, it was noted that the moment at which tobacco left supervision was sometimes delayed leading to higher Community expenditure.

When the four year time limit has nearly expired, Greek exporters place their baled tobacco under customs export warehousing control pending export to third countries and claim an advance of export refund. The Greek authorities do not, however, release the tobacco from supervision under the premium scheme when the tobacco is put under customs control. It only leaves supervision when the tobacco is actually exported (up to 6 months later). The advance of premium is cleared and the balance paid, based on the date of third country export, using the representative (green) rate for the Drachma applying on the date of export.

In this way processors have received additional payments of premium which reflect the devaluation of the green Drachma against the ECU during the period the tobacco remained under customs warehousing control. 3.35. This is an incorrect application of the regulations. While eligibility for premiums is dependent on export (or delivery to manufacture) the Court considers that this is satisfactorily met when the tobacco is placed under customs export control and an advance of refund made. The final amount of premium and any balance due should be calculated and paid based on this date. An example of the additional cost involved is the 36 469 893 DR paid to one processor in respect of tobacco put under customs warehousing control on 20 and 21 December 1990.

The tobacco was exported to third countries on 18 and 19 June 1991, immediately after a green rate change on 17 June 1991, resulting in an additional amount of 76 951 ECU being charged to the budget. 3.36. Table 6 shows an example of the comparative costs of declaring to the Commission the premium expenditure when the tobacco is put under customs control against declaring it when it is exported. EOK has provided the Court with details of eight further transactions where this procedure was followed. The approximate undue cost to the Community budget was 140 000 ECU. 3.37. Before the Court's audit took place, an EOK official had direct access to the Ministry's computer system and input data for the payment of premiums directly, using a terminal at DIDAGEP.

The risk that incorrect payments could be generated was not compensated for by other checks. DIDAGEP's subsequent check on the authorization of payments was unsatisfactory in so far as it was carried out on data recorded on the outside of envelopes provided by EOK, such as beneficiaries' names and payment order numbers, which were compared to a payment list also furnished by EOK. The envelopes themselves containing the supporting documents for each payment were not opened. Under such a system the regularity and legality of payments could not be assured. The procedure has changed now and access to the computer system is denied to non-Ministry personnel.

Supporting documentation should always be examined and payment orders should be marked with the word 'paid`. In addition, to prevent the re-use of supporting documents to create duplicate payments, these documents should also be marked 'paid` at the time of payment. 3.38. The Greek authorities have not carried out audits at processors and exporters of tobacco under the terms of Regulation (EEC) No 4045/89 (), despite the fact that significant amounts have been paid to first processors and exporters of tobacco. In 1990 and in 1991 between 45 and 49 undertakings out of a total of 71 receiving aid should have been audited at least once every two years.

3.39. The Greek authorities have said that the inspection directorate is understaffed and not yet adequately trained. However, the failure to apply the above-mentioned regulation is not acceptable and leaves Community payments at risk. Spain Table 6 - Comparative cost to the Community budget >TABLE> >TABLE> 3.40. Processing takes place in Spain at eight processing plants. In recent years processing and management of national production has undergone reorganisation. Until 1986 cultivation and first processing of tobacco was monopolized by the state body Servicio Nacional de Cultivo y Fermentación del Tabaco (SNCFT).

Since accession to the Community, controls over the cultivation of tobacco were relaxed and the processing arm of SNCFT was dispersed as CETARSA (Compañia Española de Tabaco en Rama S.A.). The manufacturing of Spanish cigarettes, however, is still carried out by the nationalized company 'TABACALERA`. Although other independent tobacco processors have entered the market, CETARSA continues to play a major role in the processing of tobacco and is the sole significant processor of the dark fermented Burley variety (Burley F). 3.41. From 1986 to 1991 ANT (Agencia Nacional del Tabaco) was the paying agency and control body for the tobacco market.

However, in April 1991, control and payment functions for this market were taken over by SENPA, the main Spanish paying agency. 3.42. The eighth recital of the preamble to the basic Regulation () sets out, inter alia, the aim that freedom of contract should be preserved, (i.e. that producers should be free to contract to sell their tobacco to whomsoever they please). The audit of the Spanish tobacco market, and of the legal relationships between TABACALERA, CETARSA and the national authorities gave rise to the following observations: (a)The Spanish Royal Decree No 573/87 established that: (i)TABACALERA would have a minority share in CETARSA (article 5.1.b); this share is currently around 20 %;

(ii)the President of CETARSA is appointed following a joint proposal of the Ministry of Economy and Finance and the Ministry of Agriculture, Fishery and Food; in addition, representatives of the Ministry of Agriculture would be members of the Board of Directors of CETARSA; (b)representatives of the Ministry of Agriculture are also members of the Board of Directors of TABACALERA, one of them was the director of ANT. 3.43. Under these arrangements, the Ministry of Agriculture, (responsible for the control of the market) also has a commercial interest in the market (via CETARSA and TABACALERA). In 1989 the connection was even closer as a high executive of CETARSA was appointed as a director of ANT.

3.44. The Spanish authorities and the Commission should review this arrangement to see whether it meets the requirements of the common organisation of the market. 3.45. Although there is a system of registering contracts in Spain, ANT/SENPA did not carry out any checks to ensure that contracts were genuine and valid. The following anomalies were observed: (a)the existence of duplicate contracts for tobacco with different processors for the same variety; (b)the existence of contracts between the same growers but different processors, where the same tobacco would be supplied as Burley E to one processor and as Burley F to the other processor.

3.46. These anomalies led to unrealistic overestimates of the quantities to be supplied which resulted in a breakdown of the control procedure. In fact, in 1991/92 82 631 tonnes were contracted, though only half of this amount was actually supplied. 3.47. The problem of carrying out checks on areas cultivated is exacerbated by the fact that there are no up-to-date land registers in Spain. Therefore, it is not possible to separately identify each plot of land in contracts by using cadastral references. 3.48. The Spanish authorities have, since the audit, addressed this problem by commencing checks on contracts, including the quantities contracted for and the field locations, introducing a computerized system of registration of contracts and holding information meetings with growers.

As a consequence, the number of contracts registered in Spain in 1992/93 dropped to 5 000 compared to 8 000 in 1991/92 (37 %) and the quantity contracted fell by 35 % from 82 631 to 55 000 tonnes. The Commission should examine the situation prior to 1992 and ensure that, where necessary, recoveries are made. 3.49. In the marketing years 1987/88 to 1889/90 premiums were paid on a total of 96 582 kg of Havanna E tobacco cultivated in the Campezo region of the Basque country, although the region was not listed in the regulation as one of the approved zones for this variety. It is clear that ANT was aware of the problem in January 1991 following a request for information from a private, multinational processor.

However, it did not take action to recover the premiums incorrectly paid. Indeed, on 4 February 1991 the director of ANT advised the processor to take Havanna from the ineligible zone under supervision 'just in case the Commission approved the zone for that marketing year` (i.e. 1990/91). The very next day he notified CETARSA to the contrary, namely to stop taking such tobacco under supervision. This letter also stated that ANT would require retrospective recognition of the zone, but again, no such request was made to the Commission nor recognition granted. Yet no steps were taken to recover the aid wrongly paid until the Court raised the matter with SENPA following the audit.

SENPA has required the repayment of 29 Mio PTA (0,23 Mio ECU). 3.50. The growing of certain varieties has expanded, for example the switch from Burley F to Burley E and the increase in the cultivation of some tobaccos in non-traditional areas. As the rule concerning eligible communes had not been complied with, the Spanish authorities were asked by the Court's auditors, during the audit, when they intended to introduce such checks and, more importantly, to carry out a review of past payments to ensure that the five year rule had been met. Although a check has now been introduced, a review of previous contracts, to provide assurance that the regulation was complied with and payments justified, has not been carried out.

The Spanish authorities maintain that the five year period could only commence from the date of accession, and therefore argue that they should only apply the regulation from 1991 onwards rather than from 1988 when it came into force. The Court believes that this is a misinterpretation of the regulation, which has led to substantial overpayments of premium which are as yet unquantifiable due to the position of the Spanish authorities. 3.51. The Court's audit confirmed the Commission's clearance auditors findings, that there is no identifiable premium certificate as required by the regulations. Instead a file of loose documents is used covering the purchases of a given variety of tobacco by a particular processor.

This file comprises nine separate books and documents. Apart from the fact that such a system does not comply with the Community's regulations, it is also cumbersome and ill-adapted to control purposes. Although in January 1991 the Spanish authorities were requested by the Commission to conform with Community regulations, the situation still persisted at the Court's visit in February 1992. 3.52. In Granada, SENPA inspectors did not have copies of the contracts between growers and the companies that they were checking in order to establish the variety and quantity to be taken under supervision. Under these circumstances it was quite possible that deliveries without contracts or deliveries substantially over the contracted quantities have occurred.

Following the Court's audit such checks have now been generally introduced. 3.53. Growers, who had a contract for Burley F tobacco with the processor CETARSA, were selected for an audit visit by the Court. It was found that most of these growers delivered tobacco which they claimed was Burley E under contracts concluded by a cooperative with another private processor. The reason for this was that they had not accepted the price offered by CETARSA and had decided to deliver all or some of the tobacco already contracted with CETARSA to the other processor offering better prices. They did this either in their own name or using a relative's name for the cooperative contract.

Some of the growers were not even members of the cooperative. The verbal promise to join the cooperative in the future was enough to permit the delivery through its cooperative contract. 3.54. As a result the total quantity of tobacco delivered by the cooperative to the first processor greatly exceeded the quantity contracted by it, which, in 1991 amounted to 265 000 kg of Burley E. The total quantity of Burley E delivered until 27 January 1992 (when the Court's audit was carried out) amounted to 418 393 kg, an increase of 58 %. 3.55. The Spanish authorities have carried out an examination at the Court's request and have decided to recover payments of aid in respect of:

(a)non-members of the cooperative; (b)tobacco delivered in excess of the quantity which may be produced on the area declared by the cooperative. To date 2 037 330 PTA (15 800 ECU) have been recovered. However, the Court considers that premiums paid on all tobacco delivered in excess of the contracted quantity should be recovered. 3.56. Another contract between a grower cooperative and a processor for 1991/92 bore little relationship to the quantities supplied, as 4 000 tonnes were contracted for but, by the end of the delivery period, only about 900 tonnes had been supplied. During the audit, the cooperative president admitted that the contract had been of a speculative nature.

Although such speculative contracts can be used as vehicles to introduce ineligible tobacco, the Spanish authorities considered that such a contract was within the terms of the regulations as the quantity delivered was less than the amount specified in the contract. Such an interpretation, if correct, would negate the use of contracts as a market regulatory tool. There should always be a close relationship between the contracted quantity and the quantity really to be supplied. 3.57. The main tobacco grown in Spain is Burley. The Community regulations recognize two distinct varieties, Burley E which is mechanically processed and Burley F which is fermented as part of the processing procedures.

When raw, the tobaccos are interchangeable but, after processing, fermented Burley is identifiable as it is much darker. It is used to manufacture dark cigarettes. Until 1993 the two types of tobacco had different norm prices and premiums, which inevitably led to control problems. The higher premiums for Burley E compared to Burley F provided an incentive for irregularities. The risk of false declaration was increased by the fact that the Spanish authorities accepted the grower's verbal declaration of the variety he delivered without checking against the contract. This led to the irregularities mentioned above.

3.58. Both the procedures carried out by ANT centrally and the physical checks carried out by the ANT inspectors in the regions were in many cases unsatisfactory. In 1991 following suspicions of irregularities, ANT decided to provisionally suspend the payment of premium advances and opened an enquiry. Because of the lack of cadastral records and the lack of supervision on the spot no effective control of the cultivation contracts was possible; after 7 months of investigations ANT did not arrive at any results, no fraud was detected and no premiums were recovered. Following the assumption of responsibility for control by SENPA, procedures have been improved and recoveries of aid made.

3.59. In 1988-89 a processing company in Spain purchased some 1 750 tonnes of Burley E tobacco, which, although taken under official control as Burley E was subsequently fermented as Burley F. Nevertheless ANT paid the higher premium for Burley E. The matter was raised with SENPA who, following a review, have requested the repayment of 375 Mio PTA (about 2,8 Mio ECU). 3.60. As it is impossible to establish a clear physical difference between Burley E and Burley F raw tobacco the distinction in the regulations between the two varieties should be removed. Although ANT/SENPA had proposed to the Commission that both Burley E and Burley F should have the same premium, no action has yet been taken.

Under the market reform the premium for 1993 has been set at the same level for both varieties but the varietal distinction still remains. 3.61. In Granada only 5 % of bales were sampled for moisture checks. The leaves from the bales were not mixed as required by the regulations and were roughly chopped using a kitchen machine which did not provide the correct sample size. Thus, the samples were not taken as foreseen by the regulations with a consequent risk of inaccurate test results. 3.62. The Spanish authorities adjust the weight for moisture content without applying the 1 % tolerance, to ensure that the growers cannot benefit by adding water to the raw tobacco leaves before weighing.

However, the formula used to adjust the weight is mathematically incorrect, as the percentage difference between the actual moisture content and the standard moisture content is simply added to or deducted from the actual weight, for example: Actual weight 100 000 kg Actual moisture content 21 % Standard moisture content 19 % Difference 2 % 'Weight` at standard moisture content = 98 000 kg. To reflect reality the correct formula for calculating the weight at standard moisture content is: 100 000 × (100 - 21): (100 - 19) = 97 531 kg 3.63. In the above example the Spanish authorities would have paid premium on 98 000 kg instead of 97 531 kg, resulting in an overpayment of premium.

However, if the leaf tobacco is drier than the standard moisture content, this calculation would under-compensate for the weight correction with implications for the minimum price to be paid to the grower. Until a correct standard formula is applied there can be no assurance of the correctness of payments. 3.64. In view of the poor checks on moisture content and the confusion between the two Burley varieties in Spain, the Court took 34 samples of tobacco from tobacco stores and growers in Spain. 14 of them were accepted by the storekeeper as representative of the tobacco taken under supervision.

The rest of the samples were taken by the Court's auditors from evidently very bad quality tobaccos in stores in Granada and Caceres. The samples were analyzed by an institute which is independent both from the Spanish authorities and from the processors. The institute was asked to indicate whether the notional variety and quality of the tobacco could be confirmed. The analyses indicated that from the 34 samples: (a)six were too wet when taken under supervision, so that the tobacco had deteriorated in quality during storage; (b)two samples were so bad that they were not eligible for premium, either as Burley E or as Burley F;

(c)one sample of tobacco taken under supervision as Virginia could not be identified as such; (d)one sample was of such a poor quality that the tobacco did not even burn due to a high salt concentration in the soil where it was grown. 3.65. This calls into question the eligibility of these tobaccos for premium although taken under supervision by ANT inspectors. In addition the institute considered that five samples of Burley taken under supervision at one processing plant as Burley E were quite similar to three others taken under supervision by another processor as Burley F (see paragraph 3.57).

Also chemical analysis did not show significant differences. 3.66. In August 1989, December 1990 and May 1991 fire destroyed tobacco in the warehouses of one processor. Advances had been paid on the tobacco taken under supervision but they had not been recovered by ANT. The matter was raised with SENPA by the Court's auditors as force majeure did not appear to apply, particularly as national legislation required processors to take out insurance covering fire risks. SENPA has recovered the amount of 679 Mio PTA (about 5,2 Mio ECU) advanced. It now requires insurance companies to submit any reports on damage claims to them and for SENPA to be the first beneficiary under the policy.

This is a procedure which should be adopted by all Member States. 3.67. In order to maintain effective control over tobacco receipts, processing and disposals it is necessary for at least one member of the control staff to be present in each processing plant. The auditors visited a processing plant without prior notice and noted that tobacco was being delivered in the absence of officials. The following day two officials arrived during the morning but left four hours later in order to catch a bus back to their office having explained that the official car had not been serviceable for three months.

This attendance calls into question the completeness and effectiveness of the controls carried out during that period. 3.68. The auditors spent a number of days at the same processor's premises. During that period tobacco refuse was frequently burnt in a pit in front of the premises. The destruction was not subject to official supervision or documentary control, leading to the possibility that the tobacco refuse could be reintroduced with other tobacco bales and thus that premium could be claimed on it and paid. 3.69. SENPA has since said that it aims to have adequate staff coverage to ensure that the processing plants, stocks and processing activities can be properly inspected and controlled during all times that the plants are open.

France 3.70. There are only three processors of tobacco in France one of which is situated in the French overseas department of Reunion. Two-thirds of all production concerns the dark air cured variety, Paraguay, which is used in the production of traditional French cigarettes. The remaining production consists of mainly Burley and Virginia with small quantities of Nijkerk and Misionero - the latter grown in Reunion. Tobacco is delivered to the processors via regional collection centres. The paying agency for tobacco is ONIFLHOR (Office national interprofessionnel des fruits légumes et horticulture), which has six regional inspectors charged with the control of growers, collection centres and processors.

Most of the tobacco processed is sold on the market with only two sales into intervention in recent years (1988/89) of Burley and Paraguay. 3.71. At the time of the audit the six inspectors working for ONIFLHOR were on loan from SEITA, the main processor and national monopoly for manufacturing tobacco. Although they were paid by ONIFLHOR, this situation was unsatisfactory, because the controllers could find themselves with a conflict of interests. 3.72. ONIFLHOR inspectors monitor yields throughout the different regions, but while quantities (which make up one part of the yield calculation) are systematically checked in the field, areas (which make up the other part of the yield calculation) are not.

Thus, there is no assurance that all of the tobacco delivered under a specific cultivation contract has really been produced on the areas indicated in the contract. 3.73. The six inspectors have responsibility for inspecting tobacco in the field, advising growers and processors, taking tobacco under supervision, establishing the premium certificate and releasing tobacco from supervision. It is evident, therefore, that they cannot be present at all the operations taking tobacco under supervision and releasing it therefrom. Until full coverage can be introduced the attendance of inspectors at collection points when tobacco is taken under supervision should be planned on a proper sample basis.

3.74. There is, however, no detailed plan governing on-the-spot visits performed by the inspectors although the general aim is to be out making visits, three days a week. The choice of where, what, when and how controls should be performed is left to the judgement of the local inspectors. Checks on the buying operations of processors were carried out without a specific sample plan. Furthermore there was no centrally approved list of minimum checks to be carried out. 3.75. As inspectors do not attend all buying operations of the processors, the premium certificates are not established when tobacco is taken under control.

They are drawn up once a week and are based on data furnished by the processor but which has not been independently checked by the inspectors. 3.76. French leaf tobacco is mostly purchased with a moisture content below the EC standard moisture content. This means that an additional amount of premium will often be owed to the processor when the weight is adjusted at the end of the marketing year. While the processor will receive any additional benefits arising from such an adjustment, he does not pass it on to the growers themselves. There is thus a risk that the actual price paid to the growers may be less than the minimum price.

While the French authorities have confirmed this to the Court, at the time of the audit the risk was small because the prices paid were significantly higher than the minimum prices. 3.77. At the two main processing centres visited the entire testing procedure is carried out by employees of the processors and ONIFLHOR does not carry out independent checks on the moisture content. Furthermore, the samples taken were sent for analysis in unsealed plastic bags being merely closed with a rubber band. There was no security to prevent the contents from being changed between the point of sample taking and the testing laboratory.

3.78. Tobacco which has come under supervision was not adjusted for the weight of impurities removed (foreign bodies and low grade tobacco) in the payment of the premium. Processors were thus paid a premium on the weight of impurities included in the tobacco purchased. While the quantities of impurities removed during processing may be small, proper assessment and adjustments should be made according to the regulations. The French authorities have agreed to introduce appropriate checks. 3.79. When tobacco is purchased, the minimum prices of the different qualities of tobacco are displayed at the purchasing centres on a board.

There is, however, no guarantee that each grower was paid these minimum prices. The ONIFLHOR inspectors check that overall the minimum price has been paid for each tobacco grade but do not check that each grower has received the minimum price or that the amount was actually paid, for example by checks on bank payments and processors accounts. 3.80. One of the processors had tobacco taken under supervision at outlying purchasing centres which was transported to the processing centre in lorries which were not properly sealed. While numbered seals were used to lock the back doors of the trucks, access could be gained through the plastic tarpaulins at the side, which were only secured with rope or rubber.

Indeed, during the Court's visit the processor's staff unloaded the trucks from the sides without breaking the seal. It is thus possible for tobacco to be substituted while the official seals remain intact. 3.81. The tobacco is re-weighed at the processing centre but it is the weight established at the purchasing centre which determines the quantity eligible for the premium, provided the weight difference is within 1 %. Small quantities of tobacco can therefore be removed undetected, as the reweighing will often take place in the absence of an inspector. 3.82. Tobacco is sent from a given processing centre to undergo stripping at another plant.

There are no controls over the weights and quality of tobacco being transferred, nor are the lorries sealed despite the fact that this tobacco is still, formally, under supervision. The inspectors merely check movements against a stock movement list. Inspectors should independently check weights, transport documents and invoices at the time of delivery and require adequate written advance warning of all deliveries. 3.83. Inspectors are advised by telephone of deliveries only shortly before the despatch is due to take place. When checking, the inspectors do not refer to transport documents, or invoices but only to the delivery notes, nor are independent checks made on the weights of boxes of tobacco and the moisture content.

Instead, information provided by the processor is used as the basis for the control. 3.84. At one of the processors, the formal exit from control is made when all tobacco of a certain variety for a certain harvest year has been sold or exported. The green ECU rate applied for the calculation of the final premium is that of the day when the last part of the tobacco has been sold or exported. In the past, up to six years have elapsed between the date that the first consignment of tobacco from a given harvest year left supervision and the date that the last of the tobacco left supervision. 3.85. The depreciation of the French franc during this period has meant that additional expenditure was charged to the Community budget.

There is no justification for this. The green rate should be that applicable at the time that each consignment of tobacco leaves control. The total amount overpaid can only be ascertained by recalculating each premium certificate. An example is the amount involved for Paraguay from the 1982 harvest which, although it had all left the processing premises by 1983, was not finally cleared from supervision until 1988 leading to an overpayment of 3 140 673 FF (about 0,45 Mio ECU). 3.86. Some assurance concerning payments made, despite the weaknesses identified in the normal control procedures could have been provided by the checks carried out under Regulation (EEC) No 4045/89 by the responsible control body (ACOFA).

However, the scrutinies carried out were, in general, unsatisfactory in that they only related to selected premium transactions rather than a detailed examination of the underlying financial records and accounts. 3.87. The scrutiny of EAGGF payments carried out by ACOFA at the processor's premises are based on documentation such as goods inward notes in the case of purchases and delivery notes in the case of sales within the Community, and shipping documents and customs proof of export and subsequent arrival where sales occur outside the Community. Stock records and, to a limited extent, production registers are also examined.

Financial records should also be examined but no checks were performed on the receipt and payment registers and the related bank statements in order to prove the validity of recorded transactions. 3.88. As the scrutinies only repeated checks previously made by the ONIFLHOR control section they had little value under the scrutiny regulation. Furthermore, no examination was made of export refund transactions. In general, the scrutinies carried out were poorly documented. In view of the amounts paid to the two main processors of about 150 Mio FF (about 21,4 Mio ECU) in both 1989 and 1990 for one undertaking and 355 Mio FF (about 50,2 Mio ECU) and 315 Mio FF (about 45,4 Mio ECU) for the other, the scrutinies performed by ACOFA under the regulation were not satisfactory.

Italy 3.89. Italy, which has some 130 processors, is the most important country for tobacco growing in the EC, as it not only supplies its own processing industry but also consigns raw tobacco to other Member States for processing. In addition, Italian processors purchase Greek tobacco for processing in Italy. The Italian paying agency AIMA is responsible for all controls necessary to implement the CMO and this is carried out through offices sited in the tobacco growing regions. 3.90. The audit in Italy can be divided into two distinct phases. The first phase examined the control system applied by AIMA until January 1992.

A number of serious weaknesses were found. 3.91. Contracts are registered in each region. The contract files examined at the regional office at Benevento contained contracts with undated or unsigned declarations of cultivation and the omission of the harvest year. One contract was stamped as received on 29.6.90 although the contract itself was dated 2 July 1990. 3.92. These omissions and errors indicate a laxness in AIMA controls. AIMA officials have said that their inspection of documents was limited and that they saw their task as formally registering documents (i.e. stamping and numbering them).

Following the audit AIMA has undertaken to ensure that checks are improved. All contracts and supporting documents will now be examined to check their validity etc. before registration. 3.93. The physical checks carried out by AIMA inspectors were also found to have major weaknesses, which stemmed from inadequate staffing of the regional tobacco offices for carrying out proper control duties: (a)there was no control programme or plan detailing the checks to be made; (b)there was no risk evaluation of processors; (c)there was no special review of cooperatives in view of the particular risks that they pose (arms-length sales do not occur);

(d)inspectors did not write a report of each control visit that they made; (e)no summary statistics were maintained by inspectors evaluating differences in weight, moisture and quality detected by them; (f)processing premises were not secure, access was available to stocks when no official was present, premises were not sealed; (g)there was no attempt to identify tobacco taken under control, by marking individual bales of leaf tobacco and keeping records of their delivery into processing and re-warehousing after processing; (h)AIMA delegated to the president of the grower cooperatives the duty to ensure that the minimum price had been received, no official check was carried out.

3.94. In addition, AIMA did not make systematic visits on the spot to the tobacco fields as part of their checks nor did they inspect reports made by the processor's own agronomists concerning field visits or indeed discuss the varieties grown with them. At the time of the audit the Guardia di Finanza was investigating many hundreds of declarations of cultivation as part of an enquiry into alleged false deliveries of tobacco to a processor in Campania. The fact that the checks carried out by AIMA were not able to identify false declarations of cultivation and the submission of irregular declarations at an earlier stage shows that their checks needed to be strengthened.

AIMA has notified the Court since the audit visit that all regional offices now carry out on-the-spot checks of cultivated areas and actual yields. 3.95. Furthermore, the AIMA inspectors did not take samples themselves when tobacco was placed under supervision but allowed processors to select them and put them to one side for the inspector to examine later. 3.96. The control was made even weaker by the fact that AIMA did not lay down minimum percentage checks to be carried out. Inspectors were left to their own devices as to if, when and with what frequency they would attend processing plants to make checks.

Furthermore, in order to carry out their duties they would telephone the processor and request that a car be provided to carry them from the office to the premises. Any chance of unannounced checks was thereby removed. Indeed the whole basis for the control rested on the cooperation of the organisation being checked. Given the allegations of misdescription of tobacco in Italy, and Campania in particular, this arrangement was clearly inappropriate. 3.97. Following critical remarks by the Court's auditors during the audit the Italian authorities revised their controls and from February 1992 contracted a large part of the physical checks to a private control company (SITRIS).

This now means that either an inspector of SITRIS or an official of AIMA is present in a processing plant whenever it is working. The control system operated by SITRIS was examined during the second phase of the audit at one processing plant and a number of weaknesses were identified: (a)one of the duties of SITRIS is to ensure that each grower delivering tobacco has a valid contract or cultivation declaration. However, no checks were made to ensure that the quantity of tobacco delivered did not exceed the quantity contracted for; (b)the quantity of leaves tested was not representative of the consignment.

Furthermore, the leaves were kept in unsealed plastic bags and stored in unlocked places, leaving the samples without any protection against substitution; (c)the samples of tobacco were cut by hand with a knife, and the cutting width was up to 1,5-1,7 cm. Tests carried out by the auditors showed that the incorrect cutting-width has a significant influence on moisture content determined; (d)the Brabender oven used for carrying out moisture tests was freely accessible during the day and distant from the hall where the inspector was taking tobacco under supervision; in spite of the inspector's evident lack of control over the testing machinery, the correct setting of the oven had never been independently checked by SITRIS or by AIMA;

(e)when calculating the correction for moisture content SITRIS and AIMA use a similarly incorrect formula as SENPA in Spain; this has been drawn to AIMA's attention but they have declined to introduce the mathematically correct formula; (f)according to the internal instructions of SITRIS the grading is left exclusively to the two representatives of the processor and the grower; in one purchasing centre the inspector made his own evaluation but it was not taken into account and was destroyed by him after the grading had taken place. 3.98. Where the processor/first buyer is a cooperative, minimum prices were not checked.

The Italian authorities believe that inside these associations the tobacco is 'transferred` from the members to the cooperative, not sold and therefore no price was established. However, several grower members of cooperatives were encountered during the audit who claimed not to have received the minimum price. AIMA has now agreed to carry out price checks at cooperatives. 3.99. In one processor's warehouse in Puglia containing raw tobacco under supervision, sacks of raw tobacco were labelled 'Jugoslavia` and 'Turkey`. The inspector said that in his opinion some Italian growers had re-used old hessian sacks, but he had no explanation as to why these sacks were labelled in this way.

Thus acceptance of third country tobacco under supervision could not be ruled out. As bales were neither numbered nor identifiable by growers name, variety, origin and lot, there was no possibility of further investigation. 3.100. Control of first processors and exporters is based on records kept by the AIMA or SITRIS inspectors in official registers outside the bookkeeping system of the traders. There is no check between these and the books of account and the financial records of the undertakings to ensure the authenticity and accuracy of the transactions entered in the registers. Therefore, as in Greece, no assurance can be given about the reality of the data on which payments are based.

Such controls are very important. Premiums have been paid in Italy on fraudulent transactions concerning non-existent tobacco. Regular checks to ensure that purchases are genuinely linked to payment and banking records are in the Court's view essential. 3.101. In the processing plants visited by the Court's auditors in Aprilia and Lecce, the security of stocks of processed tobacco is still not assured: (a)the individual tobacco bales are neither numbered nor labelled with particulars of origin, variety and lot; they are thus not identifiable, which makes checking of physical stocks and their reconciliation with book stocks impossible and stock accounts unreliable;

(b)movements of tobacco between stores and the processing lines are not independently registered by inspectors; they rely completely on information provided by the processor. 3.102. Tobacco production in Italy had been characterised by the rapid expansion of dark air-cured varieties in the 1980's. Initially the development was manifested by an increase in the production of Forchheimer Havanna and latterly by a switch to Badischer Geudertheimer - changes in cultivation caused by the attractiveness of the different premiums rather than market demand. In 1988 the premium for Forchheimer Havanna fell to 1,909 ECU/Kg whereas that for Badischer Geudertheimer was 2,534 ECU/Kg.

Significantly, the rates were not published in the regulation fixing norm prices and premiums for that marketing year () until 19 July 1988, long after the growers would have had to make a decision on the variety to cultivate. This indicated that some of the switch from Forchheimer Havanna to Badischer Geudertheimer may have been due to a 're-christening` of the variety already being grown rather than a genuine change to a different variety. 3.103. In view of this and the weaknesses found in the control system operated by AIMA during the relevant period, the second phase of the audit examined, inter alia, the basic documents underlying the payment of premium to one cooperative processor.

In particular the contracts and declarations of cultivation for growing Badischer Geudertheimer in 1988 were examined, together with supporting commercial documents such as: (a)delivery notes for young plants ('Buono di Consegna`) established when the plants are delivered from nurseries in February; (b)supply notes attached to invoices ('Bolla di accompagnamento`) established in September when the plants are invoiced; (c)invoices for plants; (d)delivery records of tobacco delivered for processing; (e)stock book ('Movimenti Cronologici di Magazzino`). 3.104. In general, it was found that plants were described as 'Havanna` or 'Piantine Havanna` on the delivery notes (Buono di Consegna), established in February, but were invoiced in September, as 'B.G.` or 'B.

Geudertheimer` on the annexed supply note (Bolla di Accompagnamento). However, most of the deliveries did not identify the variety of young plants being delivered. 3.105. There were no Bolla di Accompagnamento for young Havanna plants and no young Havanna plants were invoiced during 1988. The processor's stock book indicated the following: >TABLE> 3.106. Twenty-one of the declarations of cultivation where tobacco was supplied directly by growers to the cooperative had been accepted and registered by AIMA without containing any details of the variety to be grown and without the declaration being dated.

In addition, three declarations of cultivation had originally had the variety Havanna entered but this had been altered to Geudertheimer. 3.107. A number of declarations of cultivation relating to the 1990 harvest, which had also been accepted by AIMA, lacked a number of items of information required by the regulations such as: the variety, the harvest year, the density and the maximum production per hectare. In addition, several declarations were undated. 3.108. There were also other irregularities in relation to supplies by grower cooperatives. The largest number concerned one particular grower cooperative and are shown below: >TABLE> Thus over 80 % of the declarations were irregular, yet they had been accepted by AIMA without query.

3.109. This grower cooperative and eight growers were visited during the audit by the Court, in the presence of the Guardia di Finanza, in order to establish the circumstances leading to the changed variety in the 1988 declarations of cultivation for Badischer Geudertheimer. The growers were selected on the basis that they had either delivered tobacco to the grower cooperative and the variety on the declaration of cultivation had been changed from Havanna to Geudertheimer or, they had a contract with the processor for Geudertheimer but the delivery note for young plants showed 'Havanna`. 3.110.

Seven of the eight growers said that they had grown Forchheimer Havanna rather than the Geudertheimer indicated on the declarations of cultivation. There is no firm evidence to indicate who was responsible for the apparent irregularities but the changes took place without the knowledge of the growers. 3.111. The alterations in the declarations of cultivation were also discussed with representatives of the processor, who explained that growers had changed the variety that they grew because the premium for Forchheimer Havanna had fallen to 1,909 ECU/kg whereas that for Badischer Geudertheimer remained at 2,53 ECU/kg.

However, when it was pointed out that the new premium rates were not published until 19 July 1988, long after planting would have taken place, they could not explain the growers' apparent foresight. 3.112. The Italian authorities have pointed out that the growers themselves may be mistaken in the variety that they grew particularly as the name Havanna is used by some growers to describe dark varieties generally. They said that an examination of the leaves themselves would be required in order to establish the true variety - something that is of course impossible now. They further suggested that the difference between the description of the young plants on the delivery note by the nurseryman supplying the plants on behalf of the processor (as Havanna) and the invoice of the processor much later, was because the processor knew the true identity of the young plants.

If this assertion is true it would mean that the processor knew better than both the growers and the nurseryman what was grown, which is somewhat doubtful. 3.113. Samples of tobacco leaves taken under supervision by one processor as Badischer Geudertheimer during 1992 were analyzed by several experts on the Court's behalf. It was confirmed that the tobacco had no resemblance to Badischer Geudertheimer as grown in France or in Germany. However, it was recognized as a common variety of Italian dark tobacco with strong similarities to Havanna as far as form, size, texture and colour of the leaves were concerned.

3.114. It is clear that the problem was exacerbated by the failure of AIMA to reject altered contracts or contracts where no variety was shown, particularly during a sensitive year when Geudertheimer had a higher level of aid. Where contracts or declarations of cultivation are changed they should be rejected unless the changes are countersigned by both parties to the contract. Furthermore, in future, some contracts which have been altered should be selected for checks at the farm. 3.115. In 1990, there was a further problem of a change of variety. Estimates for the 1990 harvest were communicated in November 1990 by the Ministry of Agriculture to the Commission.

They were based on contracts registered and indicated an increase in the production of Paraguay from 2 064 t in 1989 to 10 600 t in 1990. However, the actual harvest in 1990 for Paraguay amounted to 4 661 tonnes only. 3.116. A letter issued on 28 December 1990 by the Director of AIMA explained that several of the regional offices had notified AIMA Rome that, for the 1990 harvest, growers had in error used seeds and plants of a variety different from the one that they contracted for. The letter authorized that the variety shown in the 1990 contracts could be changed to the variety that the grower claimed actually to have grown.

It was, however, a condition of the authorization that the premium for the new variety was lower than the initial one. 3.117. It appears that growers had signed contracts for the higher premium Paraguay, but had grown Hybrids of Badischer Geudertheimer (or Havanna or some other variety) instead. The decrease of 6 000 t in the quantity of Paraguay actually harvested means that some 1 500 (more than half) of the growers of Paraguay that year had used seeds or plants of a variety other than those contracted. Such a wide-spread change is difficult to believe particularly as the nurseryman visited by the auditors had said that the demand for Paraguay seeds and plants was high.

AIMA has explained that there was a shortage of Paraguay seed and it specifically checked that tobacco presented as Paraguay was in fact that variety. Italian Paraguay is not normally in high demand and its popularity with growers must have been increased by the higher premium available. However, the speculation in this variety was limited, as processed (baled) Paraguay is not eligible for intervention whereas Badischer Geudertheimer is, thus explaining the change back to Geudertheimer. Intervention storage 3.118. Table 7 shows end of year stocks between 1987 and 1992. As intervention in the tobacco market has taken place regularly in Greece and Italy almost all stocks were situated in these Member States and consisted mainly of dark varieties like Havanna or hybrids of Badischer Geudertheimer grown in Italy.

Intervention has been rarely resorted to in Germany and France and not at all in Spain. 3.119. Tobacco, which does not find a purchaser on the market, can be offered for intervention and the competent agencies in Member States concerned have to buy it into public storage providing no premium has been paid or advanced. The regulations give two options: (a)either the tobacco is offered directly by the grower as leaf tobacco to the intervention agency and is then processed on behalf of the agency which pays the quality related intervention price to the grower; for historical reasons this option is applied in Greece;

(b)or, as in the other Member States, only processed (baled) tobacco is offered by processors for intervention; in this case the agency pays the processor the derived intervention price laid down in the regulations which takes into account the processing costs. 3.120. From 1990 onwards the Council restricted intervention purchases of leaf tobacco to that grown under a cultivation contract and to cases where the contractual buyer went bankrupt under the condition that the tobacco is placed under supervision before the 15 May following harvest. Intervention in baled tobacco is limited to tobacco for which derived intervention prices are fixed.

In both cases the tobacco should not be offered for intervention later than two years after harvest (). Choice of the intervention centre 3.121. According to Council Regulation (EEC) No 1467/70 () Article 2, any offer of baled tobacco for intervention shall be made to the intervention agency for a storage centre chosen from the three centres nearest to the place where that tobacco is stored having undergone first processing and market preparation. >TABLE> 3.122. However, tobacco was found in Italy which had travelled about 2 100 km before being put in intervention storage. It was grown in Tuscany, sent to Campania for first processing and returned to Tuscany where it was stored as baled tobacco under supervision.

Afterwards, with the permission of the Italian authorities, it was moved again to premises of the same processor in Campania and then offered for intervention and stored at the intervention centre of 'Pignataro Maggiore` in Campania with whose storekeeper the offerer had close business relationships. 3.123. In another case, tobacco was in a store north of Rome when the intervention offer was lodged. Subsequently, with the agreement of the Italian authorities it was transported to an intervention centre in Campania where the storekeeper was a member of the same commercial group as the offerer. 3.124.

Tsebelia tobacco was purchased in Greece, transported to and processed in Italy, stored and then sold by the processor into intervention where it continued to be stored in his premises. At the time of the audit he was tendering to repurchase the tobacco. 3.125. Such procedures have no commercial sense at all. They only prove how beneficial the intervention system is when the tobacco is kept under the control of the same or closely related beneficiaries. In such circumstances there is no incentive for the processor to find a market for the tobacco he has purchased and processed. Intervention documentation 3.126.

Tobacco is offered for intervention in Italy in small bales (about 35-40 kg) and big bales or cartons of about 200 kg. When documenting the control work performed AIMA distinguishes between these two types, but does not retain any documentation concerning the control work for tobacco in small bales. Consequently there is no evidence of the checks executed and no basis for evaluation. 3.127. If tobacco is offered in big bales or cartons, a report is completed by the AIMA inspectors which indicates inter alia: (a)that a minimum of 25 % of the cartons were examined; (b)the identification number of each carton;

(c)the gross weight, tare and net weight of each carton; (d)the quality classification of each carton. 3.128. However, boxes are weighed in groups rather than individually and an average weight calculated. The tobacco in all the cartons is a blend of different qualities and the inspectors examine the whole lot and fix the overall percentage of each grade found. Cartons are then selected and a grade of tobacco allocated to each one to reflect the overall make-up of the lot. Therefore the quality classification allocated to a particular carton does not correspond to the real quality of the tobacco in that carton.

3.129. As a result the documentation held by AIMA does not reflect the facts nor the control work actually carried out. The lack of any documentation in the case of small bales and the poor quality documentation kept in the case of big bales mean that checks cannot be made by third parties and results cannot be recalculated. Furthermore, there is no means to ensure that tobacco has not been removed or substituted while in intervention storage. This is a real risk as intervention tobacco is frequently stored in the same premises as other baled tobacco. Security of the tobacco in intervention 3.130.

Cartons containing intervention tobacco should be marked by the intervention agency identifying it to be EC intervention tobacco. In Italy at one intervention store the stamp used to mark the cartons was held by the private storekeeper for his unsupervised use, furthermore, some of the cartons were not stamped. 3.131. In July 1991, intervention tobacco stored in Pignataro Maggiore was subsequently transferred to storage facilities in Sparanise owned by the same warehouse- keeper. The transfer was checked by AIMA on the basis of documents only, comprising: (a)a monthly stocktaking, end July 1991, at Pignataro Maggiore;

(b)delivery notes indicating the number of boxes, variety and harvest year; (c)a monthly stocktaking, end August 1991, at Sparanise. No physical checks were executed and neither the boxes nor their precise contents were identified. 3.132. In Greece, on 23 September 1988, 1 016,7 tonnes of Burley from the 1986 harvest were destroyed by fire while in intervention storage at a private trader's warehouse. While the building was insured the tobacco was not because EOK considered that the insurance premiums were too high. Although EOK did not take the necessary steps to safeguard the tobacco, it argued that 'force majeure` was applicable.

The Greek paying agency (DIDAGEP) accepted this argument and limited its action to notifying the loss of the tobacco to the Commission. 3.133. Under such circumstances DIDAGEP and the Commission should have rejected the 'force majeure` argument and established the value of the tobacco lost at the time of the fire, which should have been repaid to the Commission. As an indication of the lost value, similar Burley tobacco from the 1986 harvest was sold from intervention by tender in 1992 for 1,011 ECU/kg. On this basis the value of the tobacco destroyed was 1,03 Mio ECU. Export refunds 3.134. Although the Community imports large quantities of tobacco (528 000 tonnes in 1991) it also exports about half of its own production (210 000 tonnes in 1991) to third countries.

About 61 % of the exports concern sun cured and dark air cured tobaccos. Italy furnishes 45 % and Greece 36 % of all Community exports. 3.135. In order to enable traders to export Community grown tobacco export refunds are paid to cover the difference between the higher Community (market) prices and the lower prices prevailing on the world markets. To be eligible for export refunds the tobacco has to be of fair, sound and merchantable quality. Furthermore, the entitlement to an export refund is only acquired when the exporter has provided satisfactory evidence that customs formalities for export have been accomplished and the tobacco has been released on the market in the third country of destination.

Some varieties to some destinations, however, do not receive any refund. Low value exports 3.136. Although there were some exports of Virginia, Burley, Badischer Geudertheimer and Oriental varieties from Italy, France, Germany and Greece respectively at 'reasonable` commercial prices, there were also large exports from Italy of dark varieties at symbolic prices such as 25 LIT per kilo, as shown in Table 8. 3.137. Most of the tobacco exported had been stripped and this tobacco had received a premium and an export refund amounting in most cases to between 2,2 and 2,6 ECU/kg. Moreover, the premium is increased by a further 38,9 % to take account of the additional production costs for stripping.

For all the exports mentioned in Table 8 this additional payment alone exceeded the export value of the tobacco sold. For this outlay the grower would have received a minimum price for low grade tobacco of approximately 0,7 to 1,4 ECU/kg, the remainder would have been used to support the processing and export of almost valueless tobacco. 3.138. During 1991 a Greek processor exported 1 820 tonnes of Tsebelia tobacco of the 1989 crop to Bulgaria for the extremely low sales value of USD 18 203, i.e. 1 cent per kilo. The documents provided by the trader to the Greek authorities as proof of arrival for these exports were similar to the Annex II certificates referred to in Commission Regulation (EEC) No 3665/87 ().

However, according to Commission Regulation (EEC) No 137/90 () Annex II certificates and similar documents are, since January 1990, no longer considered as being valid proofs of arrival and genuine Bulgarian customs import entries were therefore required in this case for evidencing the release of the tobacco for home consumption in Bulgaria. These were not received. Instead, the importer provided a 'certificate` attesting the arrival of the tobacco and the transport company sent in partly illegible transport documents. As such documentation did not provide the evidence required by the regulation, DIDAGEP should not have paid refunds.

The Court requested the Greek authorities to supply details of other exports where, since January 1990, Annex II documents have been accepted as evidence of importation in third countries, no reply has been received. 3.139. The value of a 1990 consignment from Spain to Rumania was declared as being 2 PTA/kg whereas the export refund rate was 57 PTA/kg. According to the export refund claim the consignment was baled tobacco, but the official proof of arrival in Rumania certifies that the tobacco was raw tobacco leaves and not processed baled tobacco. ANT paid the refund of 4 636 989 PTA (35 900 ECU) without any query, although the identity of the goods was not clear and raw tobacco is ineligible for export refund.

3.140. The low export value becomes understandable if the tobacco exported was in fact unprocessed leaf tobacco. In its unprocessed, i.e. unstable state, the quality would considerably deteriorate during shipping. 3.141. In all such cases the tobacco had no real commercial value and the exporter's profit from such operations was almost entirely derived from the export refund. The extremely low commercial value of this tobacco also suggests that it was not of fair, sound and merchantable quality and therefore should not have been eligible for export refunds at all. Weight adjustments in Spain 3.142.

Another 704 062 kg of Virginia and 284 235 kg of Burley E were shipped from Spain to Yugoslavia. The two consignments had a declared moisture content of 12,1987 % and 12,44543 % respectively. ANT/SENPA had recalculated the weights to the standard moisture content which only applies to premiums and not to export refunds. Although proofs of arrival received were not valid as they were not signed by the Yugoslavian customs, refunds were paid on the higher adjusted weight of 718 809 kg for Virginia and 289 372 kg for Burley E. There is no regulatory basis for making such adjustments. 3.143. As SENPA automatically makes such adjustments for exports of tobacco taken under customs control for export and Spanish tobacco is normally drier than the standard moisture content, this practice leads in nearly all cases to an overpayment of refunds.

3.144. Another export refund claim concerned the export of tobacco from Spain to Rumania in 1990. Again ANT/SENPA adjusted the weight of the tobacco to standard moisture content which led to an increase in weight of 16 160 kg. The payment of export refund for the difference in quantity alone was 843 472 PTA (6 500 ECU) yet the total value of the consignment (which had a declared value of 1 PTA/kg (0,008 ECU) was only 767 934 PTA (5 900 ECU). Thus more refund was paid on non-existent tobacco than the value of the consignment itself. >TABLE> 3.145. Although the Court has drawn the attention of the Spanish authorities to the non-regulatory basis for adjusting quantities for export refund, they do not accept the Court's opinion and continue to pay export refunds on non-existent tobacco.

The clearance of accounts 3.146. The Commission's clearance auditors make regular audits in the Member States. The corrections made to the Member States' accounts for the years 1987-1990 were to the Community's advantage and amounted to 9,1 Mio ECU, 9,9 Mio ECU, 6,6 Mio ECU and 33,8 Mio ECU respectively. However, in clearing the accounts the Commission did not identify all the weaknesses noted in this Chapter of the report. 3.147. Until 1990 the Community regulations did not lay down a time limit for the processing of raw tobacco. During the clearance of accounts for 1989 in Greece the Commission auditors noted that a number of premium certificates remained open from the period 1981-86 with the raw tobacco taken under supervision not balanced against processed tobacco produced and sold.

The Commission's auditors examined an inventory taken by the Greek authorities and found that 2 000 688 kg of tobacco were apparently missing from stocks. The amount of the premium relating to this quantity was 308 380 858 DR (1,75 Mio ECU) and the security to be forfeited was 61 676 171 DR (0,35 Mio ECU). As a result of the audit, the Commission made a correction to the accounts for 1989 of the total of these two amounts, i.e. 370 057 029 DR (2,1 Mio ECU). 3.148. The Commission requested (in April 1991) that all premium files from the 1981-86 period which were still not closed should be examined and the payment of premium finalised.

The Greek authorities completed the review by 31 December 1992. As a result, a number of files were finalised and an additional amount of premium of 14,5 Mio ECU was paid to processors (see paragraph 3.2(b)). The remaining premium files lacked evidence of disposal of tobacco and the Greek authorities recovered or withheld payment from the processors concerned, of 372 762 124 DR (about 2,1 Mio ECU). This latter amount corresponds approximately to the quantity found to be missing following the EOK inventory. The amount corrected in the 1989 accounts will thus have to be adjusted. In addition, premium files in Greece from 1986 onwards which remain open should be finalised as soon as possible.

3.149. The Court could find no evidence that the Commission had taken similar action in relation to the other Member States, to ensure that premium files were properly managed and closed where possible. 3.150. Furthermore, the Commission had carried out checks on premiums and low value tobacco exports from Greece to Albania to which Greece exported 11 168 tonnes () in 1990. The Commission found that premiums and export refunds were irregularly paid on unprocessed tobacco involving about 88 % of the quantities exported to Albania in 1990. The related disallowances of expenditure will amount to 23,4 Mio ECU.

Exports from Italy to Albania amounted to 25 906 tonnes (1) in 1990 but 23 087 tonnes or 89 % did not reach its destination. The amount involved will be 10,4 Mio ECU (see also low value exports paragraphs 3.136 et seq.). 3.151. In August 1988 the Commission received allegations from a Greek trader that exports of leaf tobacco were being made from Greece and Italy to Bulgaria at the rate of 20 000 tonnes per annum. It was alleged that the leaf tobacco was misdescribed as baled tobacco in order to obtain export refunds (leaf tobacco is ineligible for refunds). The trader indicated that the time lag between the collection of the tobacco from farmers and its exportation was too short for processing to have taken place.

This information was not fully followed up by the Commission at the time. Given the current enquiries (1993) by the Commission into trade between Italy and Greece and Albania in 1990, the failure to act on the original information is regrettable. 3.152. In addition to carrying out regular audits for the clearance of accounts in Italy, the Commission had made a number of special enquiries: (a)a systems audit had been carried out in 1986 as part of the clearance audit, the conclusions of which were that is was 'an exemplary system, an almost perfect organization as concerns the application of the CMO regulations...`; it noted that there was 'an internal checking system with a review of information into great depth which proved to be a solid basis for what may be seen as an exemplary system of the European Tobacco Market`; these remarks related to the system described under paragraphs 3.90 - 3.96 in this report;

(b)a 'special clearance` took place in 1989 on the expenditure for 1988 on Badischer Geudertheimer; (c)an enquiry on the quality of the tobacco in intervention was made in 1990; (d)an enquiry into the increase in the production of Bright was launched in 1991. 3.153. The conclusion of the 'special clearance` of expenditure on Badischer Geudertheimer in 1989 was that the Italian control system could be given a favourable opinion and that the control of Badischer Geudertheimer was executed with the same rigor and accuracy as the controls in the past. 3.154. The Court cannot share the view of the Italian control procedures which the Commission's clearance auditors had in 1986 and 1989 (see paragraph 3.152 (a)).

The examination of the Badischer Geudertheimer files by the Court's auditors which are detailed in paragraphs 3.102 - 3.111, indicates not only weak procedures but irregularities, yet this same file had been examined by the Commission's auditors in the course of a clearance audit and the problems identified by the Court went unremarked. 3.155. Indeed, following the first stage of the Court's audit the Italian authorities recognized that their control procedures needed considerable strengthening and took urgent steps to improve matters by employing a private control firm under contract. The situation found by the Court was thus far from the exemplary system and almost perfect organization noted by the Commission's own auditors.

3.156. The problems of tobacco production in Spain similarly went unremarked by the Commission (see paragraphs 3.41 - 3.68). Expenditure was accepted, yet tobacco was being grown in non-approved zones, in ineligible communes, and supplied under ineligible contracts. Indeed, the problems relating to Burley E and Burley F tobacco were not identified by the Commission's auditors. The failure to apply the commune rule in Greece also went unremarked leading to irregular expenditure (see paragraphs 3.29 - 3.30). 4. THE MANAGEMENT OF THE TOBACCO MARKET 4.1. Successful management of the tobacco market, while ensuring adequate incomes for growers, requires in principle that prices be fixed which properly reflect world market conditions, that growers are given incentives to grow good quality varieties for which there is real demand, that excess production of unmarketable tobacco should be effectively discouraged, and that intervention stocks should be disposed of so as to maximise EC receipts while avoiding disruption of the market.

The Court has reviewed the Commission's performance in managing the tobacco market in the light of these criteria, which are the objectives set out for the CMO. The fixing of the basic parameters 4.2. Until the 1993 harvest, the key element of the CMO for raw tobacco was the Council's annual fixing of norm prices for each marketing year. These prices were taken into account when determining both the intervention prices for a standard quality of the raw tobacco leaves of each variety and the derived intervention prices for processed baled tobacco. In turn, the intervention prices determined the Community market prices as they set the minimum price levels for raw tobacco purchases from growers.

Most importantly the norm prices determined the main market instrument, i.e. the premiums which were supposed to cover the difference between the world market prices and the norm prices. Export refund rates which should cover the difference between the Community market prices and the world market prices for competing varieties are fixed separately by the Commission. Norm prices 4.3. As the Commission's records for the annual price fixings show, the norm prices were fixed for different varieties of leaf tobacco at the production stage. According to Regulation (EEC) No 727/70 () they should have been based on the price of the preceding marketing year and fixed at a level which:

(a)promotes the specialisation of the production according to economic structures and natural conditions; (b)presumes the rational management and economic viability of the undertakings; (c)improves the standard of living of tobacco producers by ensuring them a fair income. (d)should also take into account price cuts triggered by production exceeding the maximum guaranteed quantities (MGQ). 4.4. As the Council's price fixing decision largely followed the Commission's proposals, these should have been justified in terms of the target variables. Close monitoring of data measuring the target variables could therefore be expected.

4.5. Indeed, Commission Regulation (EEC) No 1076/78 () regrouped information considered to be necessary for the implementation of the CMO and which should be supplied quarterly and annually to the Commission by the Member States. However, none of the related files examined was complete. The lack of a transparent filing system in the period before 1990 rendered any review impossible. In the period January 1990 to April 1992 Italy and Spain frequently failed to comply with the requirements of the Regulation. Italy failed to comply with the reporting requirements of the regulation in 83 % of the cases investigated by the Court and Spain failed to comply in 75 % of the cases.

4.6. The information on the quality of the previous year's harvest and the average prices paid to the farmers was not provided at all by Spain and Italy. In 1991 Italy, France and Greece did not provide any or only incomplete information. With the exception of Germany, no other Member State indicated the original source of the information supplied and there was no evidence that the information had ever been checked for reliability or that any such verification procedures existed. 4.7. Although the files contained reminders from the Commission to the Member States, there were no signs of a systematic follow-up to obtain the required information.

Consequently much of the information used by the management committees during the year or for the preparation of the annual price proposals was incomplete or outdated. It therefore formed a limited basis for the Commission's management decisions. 4.8. As a result, the price proposals were not based on verifiable basic data and consequently did not reflect the up-to-date situation of the market. Nor did the norm prices adequately reflect the development in the farmers' incomes and the variation in the quality and marketability of EC tobacco. Instead the norm prices more or less followed general inflation, the starting point being the norm prices fixed in the first year of application of the CMO.

There was no evidence of how this initial level was determined. In 1991 there were only minor corrections to the norm prices and, in 1992, in expectation of the new CMO, prices were left unchanged. Finally the reform of the CMO for raw tobacco dropped price fixing completely (see paragraph 2.19). The intervention prices 4.9. Intervention prices were also fixed for all the 34 varieties and were set at 85 % of the respective norm prices. Nowhere could a justification of this discount be found, except that in other Common Market Organisations the same percentage between guide prices and intervention prices had been introduced.

The percentage itself remained unjustified. Premium rates 4.10. Premium rates were fixed for the 34 varieties in a way that they should in principle cover the gap between norm prices and world market prices of competing varieties. However the Commission wrote in a 'Background note on the common organization of the market in raw tobacco` dated 15 January 1990 (): 'Another characteristic of the basic regulation on raw tobacco when comparing with other sectors is that the formulation of its dispositions is very flexible. This means that we are not bound to follow a precise formula for fixing the amount of the premium.

This flexibility has enabled a stricter control of the expenses. In fact, if we had to apply a rigid deficiency payment system and had to cover every year the differences existing between the cost of the varieties of tobacco produced in the Community and those imported, the global expense would have been doubled.`The 'stricter control` seems to be contradicted by the sharp rise in expenditure which occurred between 1987 and 1991. 4.11. The flexibility mentioned was a consequence of the elements to be taken into account when fixing the premium rates set out in Article 4(1) of Regulation (EEC) No 727/70 ():

(a)past and foreseeable outlets for the sale of tobacco under normal conditions of competition on the Community market; (b)the influence of trends in prices for tobacco imported from third countries in so far as such tobacco is competitive as a substitute for tobacco produced in the Community; (c)a component designed to ensure, in the best possible manner, the guaranteed price and income and ensure the sale of Community tobacco. However, these elements were also based on the uncertain information collected under Regulation (EEC) No 1076/78 (). World market prices 4.12. As already outlined (see paragraph 4.3 et seq), no justification could be found for the norm prices fixed.

The Commission also had difficulties in determining the world market prices of tobacco, the other main element needed when fixing premiums. 4.13. The basic difficulty in determining the world market price is that there is no significant international trade in leaf tobacco. Therefore, a world market price for leaf tobacco, after curing but before first processing, does not exist. In order to come to comparable prices for setting the premium the Commission added the processing costs to the norm prices for raw tobacco. The premium thus covered the difference between the world market prices and adjusted 'norm prices`, both for baled tobacco.

4.14. Moreover, as the Community varieties are not widely produced and traded on the world market, the Commission had to identify trade in other varieties of a specified quality which they considered could, in theory, be used in manufacturing instead of the Community-grown varieties. However, it was difficult to identify representative transactions, at import, of those varieties. Therefore, when calculating the world market price of each Community variety, the Commission used the third-country prices of its analogue varieties, increased by the cost of insurance and transport to the Community. 4.15.

However, domestic prices of tobacco in third countries are unreliable. About 25 % of world production is bought by state or private monopolies at prices fixed for social or political reasons. They are often sold in barter trade or paid for in non-convertible currencies. Some prices, like those of Virginia and Burley in the USA are supported by stabilisation organizations. For a few of the oriental tobaccos there is only the Turkish market in Izmir, from which reference prices can be obtained. Export subsidies and trade restrictions also distort the 'world market prices`. 4.16. Furthermore, tobacco prices depend heavily on the quality of the tobacco traded.

Some European tobacco manufacturers confirmed that there is no international classification system rendering the qualities of leaf tobacco comparable. The USA, Zimbabwe, Malawi, Brazil and Thailand do have grading systems which have been in use for many years. Although both buyers and sellers find these systems acceptable, the Community's grading system differs more or less significantly from them. 4.17. Indeed tobacco is not a standard commodity with well defined international prices. On page 37 of its published report 'Tobacco to 1995` the Economist Intelligence Unit in London stated: 'The need for an inescapably high degree of approximation is explained in part by the world's exporting countries having no common grading system for any of the main types of tobacco; their production within each type, of tobaccos of different chemical compositions, from one region of the country to the other and from season to season; and their producing within single crops some choice leaf that at an auction without minimum prices would realise at least ten times as much as unattractive but useable nondescript grades`.

4.18. All this together limited the usefulness of the concept of 'world market prices`. Prices which were actually used in the Commission's calculations were compounded of such a wide variety of value ingredients that the price of any specific consignment could differ greatly from the average price of apparently similar ones. Processing costs 4.19. In its Special Report of 1987 (), the Court noted that the processing costs used for calculating the level of the premiums had been frozen since 1983. Despite the Commission's reply that it wanted to gather up-dated information for the fixing of prices and premiums for the 1988 harvest, the processing costs for the 1990 harvest were still frozen at the 1983 level.

A review of files revealed that the Commission had received information on processing costs only for some of the years in the period from 1983 to 1989 from France, Germany and Italy. Most of this information was incomplete and only a minor part of it was received on the detailed reporting forms issued by the Commission. 4.20. A comparison of the processing costs of 15 varieties, for which comparison was possible, with the study of the processing costs of the 1983 harvest ('Processing Costs - 1983 crop` ()) showed that for all 15 varieties the processing costs had decreased considerably since 1983.

A report on a mission of the Commission to Italy in June 1990 indicates that the real processing costs there were much lower than the figures reported by the first processors which had been used by the Commission for the calculation of the premiums. 4.21. The Commission has applied 39 % to 82 % lower processing costs when fixing the derived intervention price for baled tobacco. These lower costs are more in line with reality than those used for fixing premiums. Processing losses 4.22. When expressing raw tobacco, for which the norm prices were fixed, in equivalents of baled tobacco for which world market prices exist, the processing losses have to be taken into account.

In its replies to the Court's Special Report () the Commission considered altering the coefficients used in this arithmetic conversion. 4.23. A check on the Commission's calculations of premiums in its working papers for the years 1988-90 did not reveal any evidence that the loss coefficients had been altered or that a study of the actual loss of weight during processing was made. For the 1991 and 1992 harvest years, however, the loss coefficients for intervention tobacco were altered to a uniform rate but no documentation could be presented to support these alterations. Other factors 4.24. The other factors influencing the level of the cif 'world market prices` were not collected in a proper way either:

(a)traditionally the fob prices in third countries are expressed in USD and have to be converted into ECU in order to compare them with prices prevailing on the European market; although the exchange rates were indicated in the working papers, there was no indication in the files whether these were buying, selling or average rates and the date of the exchange rate had no clear relation to the date the working paper was finalized nor was it justified; (b)no evidence was found before the price proposals of 1990 that transport and insurance costs were ever investigated; for 1990 they were communicated by two tobacco manufacturers for some destinations only.

4.25. In spite of this complexity and the statement in the Commission's working papers that information on world market prices is collected from many sources, the Court's audit revealed that this information came only from a few sources delivered without any guidance as to the exact specifications of the information required or how to collect it. Thus the 'world market prices` used for the 1988 to the 1990 price proposals were based on information delivered by SEITA, BAT and an Italian expert, covering most but not all of the 34 varieties grown in the Community. The prices were estimates of possible increases or decreases of fob trading prices in the exporting countries in the year preceding the harvest year in question.

The qualities of the competing varieties were not specified. 4.26. Even if the information had been complete and consistent, tobacco manufacturers cannot be considered as an impartial source of information, as they are interested in high premiums (that lower their purchase prices) and which represent an incentive for them to understate the world market prices. The Commission never addressed this problem properly by, for example, checking the information received against customs import documents or purchase invoices etc. held by the manufacturers. 4.27. Finally, the method of calculating the price range of competing varieties (on which the rate of premium was based) and the total cost of processing, insurance and transport could not be evaluated as no detailed explanations were found in the working papers or in the files.

Formulae had not been consistently applied in the period up to 1990 inclusive. Neither changes in the formulae themselves nor differences between prices and premiums for different qualities were justified in the files. 4.28. The lack of transparency in the price and premium fixing was not improved after 1990, as the Commission stopped collecting information on prices of competing varieties and no longer produced the working paper. In expectation of the reform of the CMO, the Commission proposed prices and premiums against an unspecified background of budgetary restraints, production levels, MGQs and the evolution of intervention stocks.

4.29. Neither did the reform introduce more transparency to the criteria used for fixing the premiums. Regulation (EEC) No 2075/92 () did away with the fixing of prices and with the direct link between premiums and world market prices, it further simplified the system by fixing premiums not for each single variety but for groups of varieties according to curing: flue cured, light air cured, dark air cured, sun cured or fire cured. However, three of the sun cured Greek varieties - Basma, Katerini and Kaba-Koulak - have separate rates fixed for them due to their high production costs and high demand.

Also with regard to flue cured, light air cured and dark air cured tobaccos the premium rates have been set at a higher level in Belgium, Germany and France to take account of the need for undertakings to adjust to the lower premium levels applicable under the new scheme. 4.30. The Commission has said that the new premium rates were fixed in accordance with the 'Rapport du groupe tabac` () but this report does not contain any indication of the criteria to be taken into account in order to set the level of premium rates. 4.31. In accordance with Regulation (EEC) No 2075/92 (1) the premium rates and the supplementary amounts should be fixed 'for each tobacco harvest taking particular account of past and foreseeable possibilities for the disposal of the various types of tobacco, under normal conditions of competition, on the Community and world markets ().` There was no indication of what the Commission considers to be 'normal conditions of competition` in a regulated market, nor was there any evidence that the above mentioned variables had been taken into account.

In the Court's view the initial fixing of premium rates was more a test of the markets reaction to the level of support than an exercise based on objective data according to a clearly defined method. 4.32. In these circumstances it is impossible to have any assurance that the decisions on prices and premiums have been made according to the principles of sound financial management. Export refund rates 4.33. Export refunds were paid in addition to the premium for tobaccos exported to certain third countries. These refunds are difficult to justify as the aim of the premium was to cover the difference between world market prices and norm prices in the Community.

Furthermore, the fixing of the refund rate was also hampered by the difficulties in determining world market prices on a reliable basis. 4.34. Although refund rates were in general modest when compared with the premium rates, they nevertheless allowed Community processors to dump tobacco on the world market. Manufacturers confirmed that even marketable tobacco was exported at very low prices. 4.35. Under the new CMO, export refunds are no longer an instrument for balancing the internal market, although the regulation does not exclude their re-introduction as an exceptional measure in the case of serious market imbalance.

Establishing tobacco premium certificates 4.36. According to Articles 2 and 3 of Regulation (EEC) No 1726/70 () a premium certificate should be made out by the competent authorities in the Member States 'when the leaf tobacco comes under supervision, in a minimum of two numbered copies with one copy being issued to the purchaser and the other copies remaining at the disposal of the authorities of the Member State`. The premium certificate should disclose inter alia: (a)the qualities of tobacco supplied; (b)the price paid per kilogramme; (c)the date on which it came under supervision. 4.37. None of the Member States strictly applied these provisions.

In France, Greece, Italy and Spain the certificates are not made out at the time the tobacco comes under supervision and the information mentioned above is not contained. In Germany the premium certificate is established at the time the tobacco is taken under supervision but it is established by the processors. 4.38. The fact that none of the Member States complied with the rules shows that the requirements of the regulation were impractical. Either the competent body did not have the information at the time required or the body who had the information available was not the competent one. This problem continues under the reformed CMO and the Commission in conjunction with the Member States should ensure that the checks and data required are in line with commercial practices.

Sales of intervention tobacco 4.39. Unlike other Common Market Organizations, the Commission itself organizes the sales of intervention tobacco by tender. To date it has not used the other possibility given by Regulation (EEC) No 727/70 () of the disposal of stocks by public auction. It is not possible to place any assurance on the effectiveness of the procedures actually used in achieving the proper level of receipts for the EC. 4.40. Commission Regulation (EEC) No 3389/73 () gives general guidance on tendering procedures but does not provide specific instructions for the handling, recording and evaluation of the offers.

Nor had there been written internal instructions for Commission staff setting out the procedures to be followed when dealing with tobacco tenders. Making up lots for selling 4.41. The make-up of lots offered for sale in terms of: (a)variety or varieties, (b)quantity, (c)quality or qualities, (d)harvest years, is the first decision to be made by the Commission when starting tender procedures. Many of the lots offered for sale during 1988 to 1991 were mixtures of several varieties of tobacco often from different harvest years. The Commission's reasoning behind offering mixtures of varieties and harvest years was to encourage bids by offering commercially attractive tobaccos with those of much lower value.

4.42. However, no working papers dealing with these decisions could be identified. Moreover, by offering such mixtures the Commission Services had little idea of the values of the individual tobaccos put up for sale, thus leaving the door open to a cartel organising low bids from prospective buyers. Indeed, after 1990, when the Commission did not accept the prices tendered but re-offered the mixed lots for sale, subsequent bids increased. This general increase in the bids indicates that the Commission suffered losses in previous tenders. However, the Commission recognised the disadvantage of not knowing the values of individual tobacco varieties in a specific lot and, in 1990, reversed its previous policy, no longer offering mixed lots for sale, but lots of one variety.

Procedures for receiving offers 4.43. As will be seen from the following description the system was open to manipulation. Offers are handed in directly by the tenderers. Until September 1990, the offers were received at the Central Post Room (Courrier Central) of the Commission which issued a receipt. Immediately after the expiry of the deadline for the submission of tenders the envelopes were forwarded, unopened, to the competent market department and were accompanied by a typed list, signed by the 'Courrier Central` indicating the names of the tenderers and the times that the documents had been handed in.

4.44. In October 1990, a central collection point for tender documents was established in the building of the General Directorate for Agriculture (DG VI) and envelopes containing tenders were required to be delivered to a group of officials appointed by the Commission as witnesses for the opening of the envelopes. However, this new procedure did not significantly improve the security of handling offers. Although the tenderers continued to receive signed receipts for their offers, the summary lists established at the collection point were not signed by the officials receiving the envelopes. Also, the lists continued to be prepared by hand on unlined paper and unused space was not cancelled.

Thus additional unauthorised items could have been inserted without being detected. Opening procedures 4.45. The sealed envelopes containing the offers were opened at a set time by the group of officials appointed for the task. The tenders were firstly examined to ensure that the guarantees lodged were sufficient and thus that the offers could be considered. Their details were then listed ('Procés Verbal documents`). Examination of the twenty-six documents covering the period from April 1988 to September 1991 revealed the following weaknesses: (a)each of the documents consisted of three pages yet only the last page was ever signed by the witnesses present; the witnesses names were not printed alongside the signatures and it was not always clear that the signatures related to the appointed officials;

(b)blank lines were usually not cancelled; (c)page numbers were often changed or repeated; (d)the name of a specific tenderer was omitted in three cases relating to two tenders; (e)in one case the total number of tenders to be considered was omitted; (f)correcting fluid was used on three occasions to change the names of witnesses or the time of day (pre-printed on the forms) and, in one case, to transpose two names and amounts; none of the changes had been certified by the witnesses; (g)in one case, after recording the total number of offers received on the 'Procés Verbal` document, a further two offers were added to the form; in one other case the number of offers received was amended and in yet another, the name of the tenderer was altered; again none of the changes were certified by the witnesses;

(h)at the time offers were recorded, neither the envelopes nor the offer letters were signed by the witnesses as evidence that they had been present when each envelope had been opened and the offer examined. Summaries of offers 4.46. The individual offers were then summarized by lot of tobacco to be sold. The following weaknesses were found: (a)the summary lists again contained pages with unlined and unused spaces which were not cancelled; (b)although the witnesses should have signed all the documents prepared at the meeting, the summaries were often not signed by all the witnesses; in one case the summary sheets were not signed at all;

(c)in one invitation to tender in September 1990 (under Commission Regulation (EEC) No 2611/90 ()) which was witnessed by 3 officials, one tender had been omitted from the summary sheet, and was therefore excluded from the subsequent evaluation; in this case the offer price was 33 % higher than the highest tender which had been taken into consideration; thus in this particular case there is no assurance that the decision 'not to sell` was indeed correct. 4.47. Although, procedures have now been enhanced, further improvements should be made. A new lined document summarizing lots should be prepared.

It should identify the witnesses by name, and signature and indicate their division. The responsible officials should sign a statement that they have examined and checked the information contained in the summary documents. Preparing the decision 4.48. The tenders then having been checked, the tobacco market management department prepares a telex which is sent to the Member States setting out details of the tenders received, but omitting the names of the individual tenderers. The same details are forwarded to the Director General for Agriculture and to the tobacco management committee. The Commission then puts forward a proposal to be considered by the tobacco management committee.

4.49. During the period April 1988 to September 1991, there was no evidence that an independent check had been performed to ensure that the tender details agreed with the original information and the supporting documents. Such checks are important to ensure that no unauthorised additions or alterations are made once the initial opening and recording procedure is completed. Setting and recording acceptable minimum prices 4.50. Article 6.2 of Regulation (EEC) No 3389/73 () requires that the minimum prices for tobacco to be sold from intervention shall be fixed at a level so as not to hinder the distribution of such tobacco through normal commercial channels.

The elements required to be taken into account when determining the acceptable minimum price are the norm price, the premium, the costs of first processing and market preparation of the tobacco in question and the market situation. 4.51. For the tenders accepted during the period from 1988 to 1991, the minimum price was always set at the level of the highest tender. For many of these tenders no other elements supporting the determination of the acceptable minimum price were on file. For others the files showed evidence that calculations were made but the only criteria taken into consideration were selling losses and the costs of additional storage which would be incurred if the tenders were to be rejected.

Only in a very few cases could evidence be found that all the information required by the Regulation had in fact been considered. Since September 1991, calculations are performed on the basis of previous sales and market movements. Management of sales 4.52. Where tenders had been rejected and lots withdrawn and re-offered and sold at a later date, it was frequently, due to the lack of appropriate documentation, not possible to trace a specific lot of tobacco to its subsequent re-offer(s) for sale, and its eventual disposal. Different varieties of tobacco were offered together, then split and re-offered with other varieties and then re-offered yet again as a new mixture of varieties or a larger single sale of one variety.

In such cases it was not possible to verify that the rejection of an earlier tender had the intended effect of a lower sales loss to be financed by the Community budget, because the lots did not remain identical and the individual values of its ingredients were not known. 4.53. The difficulties that the Commission services found themselves in are illustrated by a sale from intervention which took place on 25 July 1990. Among the lots on offer was: A first lot consisting of: - 306 491 kg of Mavra; - 567 712 kg of Kaba Kulak and Elassona; - 163 129 kg of Kaba Kulak non-classic; - 229 988 kg of Katerini; - 252 516 kg of Burley.

A second lot consisting of: - 333 872 kg of Mavra; - 91 826 kg of Basmas. 4.54. In quality terms the first lot was much better than the second one because it contained a considerably higher proportion of tobacco in high demand. When a price of 101,11 DR/kg (0,528 ECU/kg) was offered for the latter one the offer was rejected and the lot withdrawn from sale. Nevertheless the better lot was sold for only 0,89 DR/kg (0,005 ECU/kg) more than the rejected offer for the inferior one. 4.55. Although the Mavra and Basmas contained in the lot withdrawn, were later sold as parts of other lots in a different variety composition for considerably more money, the decision to sell the higher quality lot at a price only fractionally above the price offered for the inferior lot remains questionable.

Under these circumstances the tender not accepted for the inferior lot represented a relatively better offer than that accepted for the higher quality lot. 4.56. In October 1990 another tender took place. Although there was a large increase in amounts bid in comparison to previous tenders, the Commission did not take advantage of increased prices but withdrew all the tobacco from sale. The reasons for cancelling the sale were twofold: (a)firstly the Commission services were attempting to sell all the stocks in one go; at that time Russia was interested in buying a large quantity of tobacco; in the event, the sale did not take place as the Commission services could not obtain satisfactory guarantees that the tobacco would not be re-sold on the world market;

(b)secondly, the competent market department decided to organize sales by variety in order to establish the real market values. 4.57. This decision appears to have been correct. Most of the tobacco was finally sold at higher prices which outweighed the additional storage costs for the one to two years longer that the tobacco stayed in store. The quantities sold included a large lot of Mavra and Tsebelia for a price of 0,46 ECU/kg - about 92 Drachma. This latter price underlines how out of touch with prices the Commission was when it was offering mixed lots of tobacco, as the calculation presented to the Management Committee assessed Mavra tobacco as having nil value and decisions had been taken based on that premise.

4.58. Another way in which the Commission left itself open to organized bidding by a cartel was the size of the lots offered for sale. During the Court's audit several Greek tobacco dealers protested that the size of lots, usually in excess of 1 000 tonnes and rarely less than 500 tonnes, effectively limited the number of undertakings who were able to bid. This was not only because of the sheer quantity of tobacco that had to be purchased, stored and marketed but also because of the amounts of the guarantees that had to be lodged when making bids, (the security required for making a bid for a lot of 1 000 tonnes was 339 000 ECU).

4.59. During the period January 1989 to December 1990 52 847 tonnes of tobacco were sold by 41 successful tenders. These 41 bids were made by only fourteen undertakings of which eleven came from one Italian company and represented 29 % of the total tobacco sold. Eight were from a Greek company representing 20 % of total tobacco sold. Five other undertakings each obtained 5 % of the tobacco sold and a further one 11 %. 4.60. In 1990 the Commission carried out an internal enquiry into the management of the tobacco division. During the enquiry the Commission received allegations from tobacco undertakings that a cartel had existed and been aided by the Head of Division.

If true the likely losses to Community funds were substantial. The report itself indicates that the possible loss over an eight year period could have been as high as 40 Mio ECU. 4.61. The Commission started an internal disciplinary procedure against the Head of Division concerned. However, no action has been taken to carry out adequate enquiries into the allegations of a cartel and the consequences for the Community budget. Apart from the possibility of police or similar enquiries, consideration should also be given to the taking of action under the competition policy (Article 85 of the EEC Treaty).

Confidentiality of Management Committee Meetings 4.62. It is essential that the Commission's deliberations, including those of the management committees, are dealt with discretely in order to prevent exploitation of insider information by persons who should not be a party to that information. During the audit it was found that the Greek authorities had regularly circulated detailed unofficial minutes of the Tobacco Management Committee meetings to the trade. The information circulated included not only general policy deliberations but, when sales of intervention tobacco were discussed, details of tender, names and prices.

These minutes, which were signed by the president of EOK, had a wide circulation, which included both the Union of Greek Tobacco Industries and the Federation of Greek Tobacco Exporters. This was not conducive to good management and broke the normal rules of confidentiality. Conversion to varieties in demand 4.63. One of the basic objectives of the CMO for raw tobacco is to convert the Community's tobacco production to varieties which are in demand. Demand is declining for cigarettes produced from traditional dark air cured varieties like Badischer Geudertheimer, Forchheimer Havanna, and Paraguay as well as Burley F, Maryland and sun cured Italian and Greek varieties such as Tsebelia and Mavra.

During the 1980s stocks in Italian Badischer Geudertheimer, Tsebelia and Mavra increased significantly resulting in the large-scale subsidy of exports. 4.64. To address the problem the Commission classified the 34 subsidized varieties into 5 groups by commercial demand (see Table 9). Percentage changes in the prices and premiums proposed by the Commission each year for varieties within each group were to reflect this market demand: (a)group I contains commercially sought after varieties like the flue cured Virginia varieties and the Greek aromatic sun cured varieties Basmas and Katerini; (b)group II contains light air cured tobaccos with 'temporary difficulties`;

(c)group III contains the varieties used in dark Spanish and French cigarettes, described as 'well suited to demand from Community manufacturers`; (d)group IV contains mainly oriental varieties with unreliable external markets like Xanti Yaka, Erzegovina and Kaba Koulak Classic; (e)group V contains the problem varieties such as Tsebelia, Mavra, hybrids of Badischer Guerdertheimer and Forchheimer Havanna. 4.65. Table 9 shows that it was not until 1991 that the Commission's conversion policy showed some, albeit limited, signs of success. Between 1988 and 1992 the Commission increased the MGQ's for the commercially sought after Virginia varieties by 60 %, yet production outstripped this increase reaching a level of 167 000 tonnes in 1992, which exceeded the MGQ for all Virginia varieties by 65 000 tonnes (64 %).

The production of Virginia in Germany and France did not exceed its MGQ in 1992, the overproduction results from Bright produced in Italy and Virginia produced in Spain and Greece, (overruns of 6 750 tonnes,14 %, 10 000 tonnes, 50 %, and 54 500 tonnes, 320 %, respectively). The failure of the Greek authorities to check whether Virginia tobacco came from eligible communes contributed to the general lack of control over production in Greece and it is not surprising that the MGQ's set each year for Greek produced Virginia were regularly exceeded. This control failure has had other effects, as growers of Virginia have, with Community aid, invested in the cultivation of the variety by purchasing flue curing ovens.

Moreover, under the reform package the Commission has now granted () a special aid to Virginia growers in Greece (amounting in total to 20 Mio ECU) to compensate them for the fact that they have no quota for Virginia tobacco. >TABLE> 4.66. Production of problem varieties more than trebled during the period 1981 to 1989 and, although production has declined significantly since 1989, in both 1991 and 1992 it was still substantially higher, (48 % and 13 % respectively), than the 1981 level. Badischer Geudertheimer was almost unknown in Italy before 1981 and, like Italian Forchheimer Havanna, has hardly had a market except for intervention storage and export at 'give away` prices.

On the basis that the MGQ's set each year for Forchheimer Havanna and Badischer Guerdertheimer accurately reflect the demand (), the total overproduction of those varieties in the period 1987-92 can be estimated at about 215 000 tonnes. A similar calculation for Tsebelia and Mavra shows an overproduction between 1987 and 1992 of 41 000 tonnes. 4.67. During these six years the premiums for these four problem varieties ranged between 1,3 and 2,8 ECU/kg. This overproduction has resulted in a cost in premiums of at least 455 Mio ECU for Badischer Geudertheimer and Forchheimer Havanna and 91 Mio ECU for Tsebelia and Mavra, amounting to a total of 546 Mio ECU.

This latter amount is a minimum as the Council has set guarantee thresholds and processing quotas for 1994 () which indicate that market requirements for these varieties are about 35 % less than the MGQ's for 1992, the last year of their application. If the 1994 quotas are taken as the basis for the calculation then the overproduction of the problem varieties during those six years has cost the Community 490 Mio ECU and 313 Mio ECU respectively, i.e. a total of over 800 Mio ECU. In addition to this cost, are the millions of ECU paid for intervention storage and export refunds resulting from the surpluses.

This expenditure could, to a large extent, have been avoided had the Commission taken effective steps at an earlier stage to restrict the increase in the production of these unwanted varieties. While it is true that the Commission have used the MGQ's to guide production and have been instrumental in the introduction of cuts in prices, premiums and export refunds for the problem varieties, the success of the Commission's management can be measured by the fact that the production of Italian problem varieties is still 34 % above the quota set for 1994 and the Greek problem varieties are 133 % above that level.

4.68. Since 1992 the Commission has initiated various conversion schemes: (a)a special programme () for conversion of the cultivation of the varieties Tsebelia, Mavra, Forchheimer Havanna IIc and hybrids of Geudertheimer into other agriculture crops for the 1993 agriculture year; (b)a diversification programme for flue cured tobacco producers in Greece, i.e. producers of Greek Virginia tobacco (); (c)financed a study by the French tobacco company SEITA on the reduction of the tar content in some specific varieties; (d)in mid-1993 steps were taken to set up a Community fund for tobacco research and information in implementation of Article 13 of Council Regulation (EEC) No 2075/92 (); the fund, which will be financed from the proceeds of a deduction not exceeding 1 % from the premium, will be used to orientate Community tobacco production towards the least harmful varieties and qualities.

4.69. The Commission's efforts to convert tobacco production into production of less harmful varieties are rather late, especially if the aim was to meet the change in demand as a consequence of the Council Directive () for low tar content in cigarettes. In addition to this the Commission has not yet financed a 'study on the possibilities for putting tobacco-growing areas to other uses`, although there are provisions for such a study in the 1990 to 1994 action plan in the context of the 'Europe against Cancer` programme (). Promotion of quality 4.70. The Court noted in the Special Report of 1987 () that the quality of Community grown varieties needed to be improved and that, due to low quality, destruction of certain varieties with compensation would have produced a budgetary saving without a financial loss to the producer.

Maximum yields are laid down for each variety in order to ensure that quality is not affected by high yields resulting from, for example, high plant density, over-irrigation or excessive fertilization. In 1992 the Commission stated in its Regulation No 1210/92 () fixing the characteristics of each variety grown in the Community that 'an increase has been observed in the yields for certain varieties since 1987 as a result of progress in production techniques without any corresponding deterioration in quality`. With the exception of Beneventano and Forchheimer Havanna the regulation allows the yields of the different varieties to increase quite substantially.

The increases are on average 34 % as compared to the maximum yield in the years 1988-91. The variety Paraguay grown in Italy which substantially exceeded its MGQ between 1988 and 1991 inclusive and which was almost entirely classified in the lowest quality grade accounts for the highest increase, 86 % (1). 4.71. In order to monitor the quality of each variety produced in the Community, the Member States are obliged, according to Regulation (EEC) No 1076/78 (), to assess the tobacco when delivered to first processing and to forward information on the average quality classification for each harvest.

However, an examination of the statistical files of the Commission, revealed that the main producing Member States rarely forwarded this information and, if they did, it was only with considerable delays. 4.72. In addition, the information provided gives no clear evidence of a general improvement in quality. The share of German varieties which can be classified as the best quality (A-quality) has either remained unchanged or decreased from 1986 to 1990. Italy did not provide any information for the harvest year 1990, but between 1986 and 1989 only Burley and Maryland improved while the quality of Bright and Kentucky remained unchanged.

For the remaining varieties none of the quantities delivered could be classified as A-quality. In Greece four varieties improved in quality from 1986 to 1990 while four others decreased. No conclusions can be drawn regarding France and Spain as these Member States have only provided information for 1990 (). In general the information provided cannot substantiate the conclusion that the quality of Community grown tobacco has improved. 4.73. This is also demonstrated by a comparison of prices (unit values) of tobacco imported and exported. Table 10 shows that the value of Community produced tobacco continues to be much lower than the price of imported tobacco of the same type.

The exceptions (fire cured tobacco other than Kentucky, and light air cured tobacco other than Burley and Maryland) are based on export of insignificant quantities. The figures must be treated with caution as they refer to tobacco types rather than varieties and the Commission Services have pointed out that imported tobaccos may have different uses than similar Community grown varieties (for example to add taste to cigarettes produced from Community grown tobaccos). 4.74. The value of Community tobacco exported to third countries is extremely low when compared with the aid paid by the Community.

The system was functioning at its worst in 1989. In that year the total aid granted in the form of premium and export refund for each kilogramme exported of the variety Hybrids of Badisher Geudertheimer was 2,97 ECU. This calculation takes into account that 1,22 kg of leaves of this variety are necessary to produce 1 kg of baled tobacco thus, 2,63 ECU were paid in premium and 0,34 ECU in export refund. The total aid was more than seven times the average export value shown in Table 10. However, the quality which was actually exported was so low that, as shown in Table 8, its value was well below the average export price of 0,4 ECU per kg.

Growers of such a low quality would have received a minimum price of 1,64 ECU or about half of the total aid paid. 4.75. Although the situation has improved somewhat since then, by a reduction in premiums, the high level of Community support is still out of proportion to the value of Community tobacco on the world market. While the export values of Bright (Italian Virginia) and Burley E in 1991 were about 1,0 ECU/kg and 0,6 ECU/kg, respectively, the total amounts of aid paid for the leaf tobacco used were 2,63 ECU and 2,33 ECU being nearly three times and nearly four times their respective export values.

4.76. Italian processors have confirmed that there was no general improvement in quality when they informed the Court's auditors that large quantities of tobacco were of such a poor quality that the only markets for it were those in Northern Africa and Eastern Europe. In the system as introduced, the premium was reduced.The Commission's Regulation (1) increasing the maximum permissible yields has not contributed to the improvement of the quality of Community grown varieties. >TABLE> 4.77. Round Scafati tobacco is cultivated in Spain. The two biggest growers of this variety are in the region of Leon.

Although the growers follow the cultivation requirements set out in the regulations, the quality of the tobacco produced is extremely poor. A SENPA internal report states that the climatic and agronomic characteristics of most of the region are not suitable for this variety. Indeed, the cultivation requirements contained in the regulations are inappropriate. For example, in order to produce tobacco of reasonable quality in the sunny climate in the plain of Leon, the plants would have to be covered most of the time by a gauze netting to protect them. Round Scafati was grown originally in small plots in cooler, drier mountain areas of the region and is not suited to mass cultivation in irrigated plots.

The Commission should have identified the problem and proposed changes to the cultivation requirements and the areas allowed for this variety. The result of the Commission's inaction has meant that premiums have been spent on producing tobacco of negligible commercial value. Limitation of production Budgetary stabilizers 4.78. A system of maximum guaranteed quantities (MGQ's) was introduced in 1988 in order to stabilize the expenditure on the CMO for raw tobacco. However the system was initially not imposed with sufficient severity to oblige growers to change their behaviour, and in some cases the small reductions in premium consequent on over production resulted in growers still further increasing their crops to offset any income loss.

In the system as introduced, the premium was reduced by 1 % for each percent that the MGQ for a certain crop was exceeded. However, the maximum reduction in the premium was limited to 5 % for the 1988 harvest. This provision was later repealed by the Court of Justice as planting had already taken place when the regulation was adopted (). For the following years the maximum rate of reduction of the premium in the case of production exceeding the MGQ was increased to 15 %. For the 1992 harvest this reduction was further increased to 23 % () additionally, in order to limit the production of low grade tobacco, if the quantity of the lowest grade of a given variety purchased by a processor exceeded a fixed maximum rate (20 %, 30 % or 45 % depending on the variety) the premium would be reduced by 30 % ().

However, the reductions in premiums, triggered when production of a given variety exceeded its MGQ, were not consolidated, either fully or in part, in the premiums fixed for the following marketing year. Ignoring those varieties that are grown only marginally, there were no reductions in premium in 1989, only three in 1990 and 13 in 1991 when there was, for the first time, a general attempt to bring expenditure under control by cutting premiums. 4.79. In the years 1988 to 1992 the total production exceeded the total MGQ's by 3 %, 7 %, 9 %, 10 % and 5 % respectively (). However, the total expenditure on premiums was not reduced by the corresponding percentages because the overproduction was concentrated on only a few varieties.

For example, Italian Badischer Geudertheimer exceeded its MGQ by 1008 % in 1989 but the reduction in premium was only 15 %, which was the limit set for that year. 4.80. The Commission used the MGQ stabilizer mechanism as a market management tool to manage trends in production, e.g. by substantially raising the MGQ's for Group I varieties in 1989. The result of this and the additional increases that were made in 1990 and 1991 without an associated reduction in premiums have been a major factor in the 53 000 tonnes overproduction in 1992 (see paragraph 4.63 et seq.). Indeed, it was not until 1991 that a general attempt was made to bring expenditure under control by cutting premiums.

Although production increased in the six year period from 1987 to 1992 by a total of 7 %, expenditure over the same period increased by 53 % (from 804 to 1233 Mio ECU). The Commission should have taken effective action earlier to bring expenditure under control e.g. by reducing premiums on varieties where production outstripped demand. The effect of rates of exchange 4.81. Before accession, cultivation contracts were not used in Greece. When Greece entered the Community this was taken into account in so far as the existence of cultivation contracts was not made obligatory (as in the other Member States) for the processor to receive an advance on the premium.

Instead an obligation was introduced that the tobacco had to be processed in Greece. In 1986 the Commission removed even this obligation with its Regulation (EEC) No 1791/86 for tobacco harvested from 1985 to 1988. This opened the door to the speculative purchase of poor quality Tsebelia and Mavra by Italian processors to benefit from green rate divergences. 4.82. Indeed, between 1987 and 1990 Tsebelia leaf tobacco was purchased in Greece by Italian processors, transported to Italy, placed under supervision and processed there. The fact that the baled tobacco was sold into intervention after processing demonstrates that these imports of a variety generally in low demand were not normal commercial operations.

The trade was induced by the divergence between the market rates and the representative (green) rates for the Drachma and the Lira. Both the minimum purchase price and the premium are fixed in ECU. However, the processors obtained an additional benefit of 15 % to 20 % from the difference between the cost of the tobacco purchased in Drachma at market rates and the premium received in Lira at the green rate of exchange. 4.83. As a consequence of this artificial demand production initially rose and then did not fall, despite cuts in premiums, to its true level of market demand of about 12 000 tonnes.

The production of Tsebelia rose from 25 400 tonnes to 30 000 tonnes during the period 1986 to 1989 and intervention stocks from the 1987-1989 harvests amounted to nearly 30 000 tonnes in 1991 based on an artificial demand of about 5 000 tonnes per year. The result was that about 38 Mio ECU were unnecessarily disbursed on premiums and the conversion of Tsebelia to other tobacco varieties was delayed. There were, of course, additional budgetary costs for those quantities delivered to intervention. The Commission failed to draw the attention of the budgetary authorities to the problem and did not suggest any means of stopping the speculative trade.

The farmers' income 4.84. An aim of the CMO is to provide the farmers with employment and fair living standards. Thus, the Commission's management should be based on regular and reliable information on the elements determining the tobacco farmers income. 4.85. The lack of information on the cost of tobacco growing was already criticised by the Court of Auditors in its Special Report No 3/87 (). In its reply the Commission said that it had undertaken to present a report on production costs in the various producer countries when making the price proposal for the 1988 harvest. A study was actually started in 1988 but did not result in any official report due to the lack of reliable information from the farmers on the cost elements, in particular regarding the remuneration for the family labour.

4.86. An examination of data furnished by the Commission's own Farm Accounting Data Network (FADN) revealed that it is rather difficult to come to meaningful conclusions (e.g. that the family labour cost in Greece in 1990 should be 266 616 ECU per hectare). The only income concept which can be used with some certainty is a net income concept which includes the remuneration of the farm labour, rent and interest. It is roughly the value of the total tobacco production after deductions of production cost (specific cost, overheads and depreciations). 4.87. For France, Italy, Greece and to some extent Spain, which together accounted for 94 % of all EAGGF expenditure in the tobacco sector between 1988 and 1990, the sample is large enough to be representative.

The following tobacco production data in relation to gross income, costs and net income are based on farms in these four Member States which have tobacco as the principal crop. They are the average figures per hectare for the years 1988-1990: >TABLE> 4.88. Some care must be taken with the averages as they are influenced by the transition of Spain to Community market support levels and by the fact that they do not include farms where tobacco growing is not the major cultivation. The variation in the gross income from the tobacco produced for the different Member States is due to differences in the values of the varieties grown, the yields, the qualities and the premiums.

Net income is also affected by these factors and by the differences in the cost of production between the varieties and the different costs in the Member States. The estimated average net income of 3 379 ECU/ha is supported by another study () made by the tobacco industry, which shows that the income from tobacco growing can be estimated at 3 151 ECU/ha in 1990. 4.89. The high gross income per hectare from growing tobacco is a direct consequence of the high level of Community aid for tobacco growing. The total EAGGF expenditure per hectare, including export refunds, intervention costs and premiums, averaged 4 802 ECU in the four Member States between 1988 and 1990.

Of this the expenditure for premiums alone averaged 4 235 ECU/ha (). These amounts correspond to 77 % and 68 %, respectively, of the grower's gross income of 6 233 ECU/ha. 4.90. Although in the period 1988-90 Community expenditure per hectare for tobacco was already very high (4 802 ECU/ha), it increased by 29 % to a level of about 6 200 ECU/ha in 1991 (2). This is some nine times higher than the support for oilseeds, twenty-one times higher than that for wine and twenty-three times higher than that for cereals (). 4.91. Tobacco growing in the Community is very costly. However, despite an average cost of production of 2 853 ECU/ha in the four Member States between 1988 and 1990 (excluding remuneration of labour, rent and interest), the tobacco growers' net income per hectare is still substantially higher than for other crops.

In comparison, the net income was more than three times as high as the average net income of 990 ECU/ha for all types of farms in the same Member States and years and was more than seven times as high as the average net income of 441 ECU/ha for farms specializing in the production of cereals (). 4.92. Although the income is high per hectare compared to other crops, other FADN data shows that the average net income for tobacco farms was 5 702 ECU compared to 9 138 ECU for other farms, i.e. 38 % less (). This reflects the fundamental problem in the tobacco sector of farm size, which averages only 1,1 hectares.

4.93. FADN also provides figures on the rent of land which on average was 724 ECU/ha for the four Member States in the same period. Thus, after deduction of this cost the average income was 2 655 ECU/ha. If any farmer not growing tobacco in the three years 1988 to 1990 had received Community compensation in lieu of the average net income foregone per hectare of 2 655 ECU, the charges to the Community budget would have been 595 Mio ECU in total. The Community could thus have saved 520 Mio ECU per year on this scheme or more than 46 % of the average annual expenditure of 1 115 Mio ECU. Such a scheme would have the added merit of reducing the production of tobacco in line with the Community's efforts on health grounds to discourage smoking.

5. HEALTH AND OTHER COSTS OF TOBACCO CONSUMPTION 5.1. Any evaluation of tobacco support schemes should take into account the health costs of tobacco consumption. Medical experts recognise tobacco use as the most important preventable cause of lung cancer, one in four smokers prematurely dies of lung cancer, coronary heart disease or chronic lung disease. In an answer to a parliamentary question the Commission stated (): 'According to WHO statistics, it is estimated that tobacco is responsible every year for the deaths of 431 000 people in the Member States of the Community. Tobacco probably now accounts for at least 25 % of all EEC deaths in middle age (35-69) and for at least 10 % of all EC deaths in old age.

Those who are killed by tobacco in middle age lose on average about 20 years each of life expectancy. The predicted annual death toll from tobacco in the major European region, which serves as a basis for WHO's estimate among those aged under 25 in 1990, is about 2 million for the year 2025...`5.2. The WHO identifies tobacco as responsible for about 90 % of deaths from lung cancer, 75 % of deaths from bronchitis/emphysema and 25 % of deaths from ischaemic heart disease (). Maternal smoking during pregnancy is associated with a higher risk of premature birth, low birth weight and stillbirth. Studies in the USA indicate that smoking retards wound healing, extends the recovery period after surgery and causes tooth losses and bone erosion ().

5.3. Furthermore, the dangers of passive smoking are becoming more evident. People regularly exposed to other people's smoke have a 10-30 % higher risk of getting lung cancer than do other non-smokers. Other research shows that nicotine, like cocaine and heroin, is addictive, psychoactive, used compulsively and repeatedly, and associated with a withdrawal symptom (). 5.4. According to the Surgeon General of the USA, although annual excise duties amounted to 22 000 Mio USD, the annual direct health costs were 39 000 Mio USD and, when including auxiliary costs, the total cost to (USA) society of smoking is 64 000 Mio USD () per year.

The Office of Technology Assessment of the US Congress has, in a comprehensive study, estimated the annual health care cost of three main groups of tobacco related diseases, i.e. cancer, and diseases of the circulatory and respiratory systems, to between 11 500 Mio USD and 34 500 Mio USD with a middle estimate of 22 000 Mio USD. Smoking related productivity losses were estimated at between 26 900 Mio USD and 60 500 Mio USD for the same disease groups with a middle estimate of 43 000 Mio USD. The total of the middle estimates being 65 000 Mio USD corresponding to 2,17 USD per packet of cigarettes sold in 1985 ().

These costs were based on the risk factors of 1982 and the price level of 1985 are understated as they do not include all smoking related diseases. 5.5. The cost of smoking related fires, cleaning etc. should also be taken into account. Indeed, valuable property and natural resources are lost annually due to smoking related fires as between a quarter and a third of all fires are caused by burning and smouldering cigarettes (). 5.6. In addition to the personal suffering associated with death and illness caused by tobacco there are other costs to society. Smokers generally have greater illness and absenteeism than non-smokers and those dying prematurely are in general trained, skilled and mature workers (3).

This implies additional medical costs and welfare costs for supporting the families which have lost their wage-earners from a smoking related illness. Anti-smoking action taken by the Commission 5.7. In June 1985 the Council approved a European programme against cancer. The programme, which has an overall target of reducing mortality from cancer by 15 % before year 2000, has been running since 1987 with two action plans (1987-89 and 1990-94 ()) taking as its first priority the campaign against tobacco. This has not only involved legislative measures but also a large-scale campaign of public awareness.

5.8. Priorities were established to reduce the impact of tobacco on health. Seven proposals for directives have been presented by the Commission: - harmonization of indirect taxation on tobacco products (three proposals); - harmonization of the health warnings on tobacco packets and banning of certain forms of oral snuff (two proposals); - limitation of the tar yield of cigarettes; - banning of tobacco advertising. All these measures, except the ban on advertising tobacco products, were adopted and have been implemented or are under implementation in all Member States. Moreover, in 1989, the Ministers of Health adopted a resolution concerning smoking in public places.

5.9. Since 1988, under the two action plans, the EC programme has financed some 150 projects relating to prevention of tobacco consumption by health education at schools, information and awareness of the general public in particular young people, and by helping people to quit smoking. 5.10. The Commission has set up a European group of anti-smoking organizations bringing together non-governmental organizations and senior officials with responsibility in this field. The programme also finances and publishes surveys and opinion polls which provide a basis for monitoring trends in opinion in the Member States.

5.11. Both action plans contain proposals for the reorientation of tobacco production towards less toxic varieties and a study of the possibilities of conversion. The second action plan stated: 'Quite clearly, the policy pursued in this sector will need in future to take more account of the health aspects;` but it also stated: 'Any policy which aims to reduce, in the short run, the European production of tobacco will also have an impact on employment in those areas already affected, as well as an increase of the trade deficit which is already relatively large. `5.12. The current action plan has a total budget of 50 Mio ECU of which the 1992 budget was 13,5 Mio ECU.

The sub-budget for tobacco smoking prevention in 1992 was 1,7 Mio ECU only. The budget for 1993 is about 10 Mio ECU with a sub-budget for tobacco smoking prevention of 1,5 Mio ECU. 5.13. The Community is co-operating with the United Nations in smoking prevention through the World Health Organization, Regional Office of Europe, but the budget for the 'Tobacco or Health` activities carried out by this organization is as low as 135 000 USD (118 000 ECU) although the activities concern 31 countries including Eastern European countries. 5.14. An evaluation of the 'Europe against Cancer` programme 1987-91, was made on behalf of the Commission in May 1991 and included a report on the prevention of tobacco consumption in terms of legislation and activities undertaken ().

This stated that the target of reducing the number of cancer deaths by 15 % before the year 2000 is unattainable unless a considerable effort is made. The report noted an increase in production of varieties with a lower tar yield, but that there had been no reduction in the production of other varieties. 5.15. Meanwhile advertising for smoking continues despite the smoking-related deaths and serious diseases. The modest sum that the Community has allocated to anti-smoking campaigns is in stark contrast to the advertisement budgets of the tobacco industry in the EC, estimated at 446 Mio ECU in 1988 ().

Implication for tobacco production 5.16. The budget of the 'Europe against Cancer` campaign and the sub-budget for smoking prevention constitute only about 1,1 % respectively and 0,1 % of the total expenditure for tobacco growing in 1992. The most important element to date has been the directive on a reduction of the tar content to 15 milligrams by the end of 1992 and to 12 milligrams from the end of the year 1997 (2000 and 2006 in the case of Greece). The directive also has an implication for nicotine which is found in direct relation to tar in the proportion 1:10. Based on 1987 figures, the industry has estimated that the directive will affect 46 % of cigarettes marketed in the EEC and 83 % at the more rigorous 12 milligrams level.

5.17. In support of the general effort to convert to lower tar varieties and to encourage tobaccos in higher demand, as part of the new tobacco reform, the Commission has initiated a three year conversion programme for Mavra, Tsebelia, Forchheimer Havanna IIc and hybrids of Geudertheimer. Farmers who convert to other crops or cultivate other varieties are compensated for their loss of income due to such conversion. 5.18. However, the effect on the conversion of production from high tar varieties to low tar varieties is not evident. Lower tar content cigarettes have been partly achieved by the looser packing of tobacco in cigarettes.

In order to give a more meaningful basis for comparison, the tar and nicotine content would need to be indicated in relation to the total content of tobacco per cigarette. Taxes 5.19. The Finnish Board of Health carried out a detailed cost/benefit analysis of the effects of an increase in the price of manufactured tobacco products. It found that the price was the most important determinant of demand and that the reduction in smoking caused by raising prices would create a net benefit to society in general (). 5.20. A possibility to raise prices is to raise sales taxes. Indeed, three different sales taxes are levied on manufactured tobacco in the Member States: - a specific tax which is a fixed amount per cigarette regardless of size or retail price, - an ad valorem excise tax which is calculated as a percentage of the excise taxes and V.A.T. inclusive recommended retail price of the product, and- V.A.T. which is determined in the usual way and is payable on the price, excise tax inclusive.

5.21. According to the directive on harmonization of indirect taxation on cigarettes the Member States should apply an overall minimum excise duty (specific plus ad valorem without VAT) of 57 % of the retail selling price of cigarettes of the most popular price category (). For other tobacco products the Member States should apply a proportional tax on the retail selling price or a specific tax where the level of taxation depends on the product (). 5.22. It has been estimated () that the total revenue from tobacco excise duties and VAT in the Community amounted to more than 36 000 Mio ECU in 1990 and thus contributed to more than 3 % of total central government tax receipts.

In addition, governments received corporation taxes from tobacco manufactures and income taxes and social security contributions from those employed in the industry. This amounted to approximately 3 000 Mio ECU in 1990. 5.23. According to studies made in the USA a 10 % increase in the prices of cigarettes will reduce the consumption among adults by only 4 % (). For teenagers with a shorter period of addiction, a less intense daily consumption and less economic means, the decrease would be 14 % with the major part of the reduced demand representing decisions not to start smoking or to quit (). These studies and data in the 1989 Surgeon General's report, support the conclusion that taxation is a powerful tool to discourage children from smoking and to encourage young adults to quit.

Exports to developing countries 5.24. The subsidising of tobacco production and export is not only an endorsement of a policy which is against the Community's own health policy but also counteracts the cooperation with developing countries. It also demonstrates the double standards of Community policy in this field. 5.25. Subsidies are paid for producing tobacco which has practically no market in the Community. Almost all this tobacco is exported to Central and Eastern Europe and North Africa where there are insufficient controls of tar content and where the countries can hardly afford to cope with additional bought-in mortality and high health costs.

5.26. The single tobacco trade balance is not only an economic issue, but also a question of whether the recruitment of developing countries to replace the shrinking Community market is compatible with the cooperation policy of the Community and morally justifiable considering the accompanying economic and health problems imposed on these countries. 5.27. The World Bank has recognised () that tobacco-related diseases disproportionately affect the poor, who are among the most difficult to reach and persuade to cease smoking. Therefore, the Bank encourages, through its activities in the health sector, the reduction of tobacco smoking and has decided not to lend money directly (and as far as possible also not indirectly), for tobacco production, processing, imports or marketing of tobacco-based products.

The above policy objective is to reduce the considerable costs, notably in medical care, entailed by tobacco consumption. However, the Community has not adopted a similar policy in relation to aid to developing countries. 6. TOBACCO IN THE USA 6.1. Public expenditure on tobacco subsidies in the USA amount to a mere 1,7 % of Community CMO expenditure while producing at the same time twice as much tobacco. Background 6.2. The dangers to health associated with the use of tobacco have had a marked effect on public opinion in the USA with cigarette consumption in particular showing a steady decline.

The health factor has also been recognised by the US legislators, in 1982 the US Congress introduced the no net-cost programme which removed direct subsidies from tobacco. Instead, market support is based on production quotas for growers and minimum prices, with tobacco that cannot be sold at or above the minimum price being purchased by producer associations with the aid of government loans. The minimum prices are conditional on producer association members approving the overall quota level by regular referenda. 6.3. As a result of the no net-cost law, no costs, except general administrative costs, can now be borne for crops grown since 1983.

There are charges by the United States Department of Agriculture (USDA) marketing inspectors for tobacco grading and testing for pesticide residues. The United States suspended its five cents/lb export subsidy in 1973. Since then, there has been no export subsidy for tobacco. 6.4. The current cost of tobacco support is in the region of 30-35 Mio USD composed of: (a)administrative costs (staff salaries, travelling and subsistence); (b)export promotion; (c)part of the cost of the Agricultural Research Service; Tobacco is also included in an export credit programme that provides for a government guarantee on commercial loans made to foreign buyers to purchase US agricultural products.

Quotas 6.5. The quotas are set by the USDA for each variety using the following information: (a)the purchase intentions of the tobacco manufacturers (); (b)the average tobacco exports over a three year period; (c)the 'stabilization` stock level, which is set at 100 Mio lbs or 15 % of the previous years quota. The inclusion of stock levels in the calculation is the key to avoiding over-production and thus a build-up of stocks. 6.6. The quotas are allocated to individual growers via the Farm Bureaux which exist in each county. If a grower markets less than the quantity he has been allocated, the difference is added to the farm's quota for the following year.

Similarly, marketings in excess of the quota are deducted from next year's quota. Production in excess of 103 % of the quota is penalised by requiring the grower to forfeit 75 % of the average market prices of the previous year. 6.7. Quotas are transferable; Virginia quotas may be sold within county boundaries, those for Burley may be leased but not sold. These restrictions are applied in order to ensure that tobacco production does not move from traditional growing areas and aim at conserving quality, local farming expertise and jobs. Without this there would be a tendency for growing to shift to areas where farming conditions are generally easier, in terms of land or irrigation.

6.8. One of the problems of the quota system identified by the USDA is the difficulty of ensuring that growers declare their actual production. Undeclared tobacco, produced over and above the quota, has an effect on the market's ability to absorb the tobacco grown. Consequently the US authorities penalise growers and dealers for non -declaration and dealers are required to lodge a bond to ensure that penalties are paid. Minimum Prices 6.9. Price support is operated through 11 producer associations under contracts with the Commodity Credit Corporation (CCC), a department within the USDA. Each year the USDA sets an average minimum price for each variety of tobacco.

These are then translated into minimum prices for each grade. The average price is set having regard to the level for the previous year, adjusted by changes in a 5-year moving average of market prices (two-thirds weight) and changes in the index of the cost of production of tobacco (one-third weight). 6.10. Most tobacco is marketed by auction, and the minimum price for each of the many grades (there are 153 for Virginia and 113 for Burley) becomes a pre-established bid by the producer association on each lot of tobacco offered. If commercial bids for any lot of tobacco are not higher, the association becomes the buyer at the minimum price.

The association pledges the tobacco as collateral and borrows funds from the CCC to pay the producers the support prices and to cover the costs of processing, storing, and selling the tobacco. 6.11. Farmers can forfeit without penalty the loan collateral (the tobacco) to the Government as settlement of the loan. However, from 1982 the No Net-Cost Act requires that all principal, interest, and other costs must be covered by proceeds from CCC sales of tobacco, with any outstanding cost met by assessments on growers and purchasers. Several years are usually required to completely dispose of a year's tobacco crop.

The loan remains outstanding as long as any of the crop remains unsold but is reduced as each sale is made. When the tobacco received from any year's crop is sold for more than enough to repay the CCC loan with interest, the net gains are set off against the losses from other years. 6.12. The cost to the grower of running the support programme has varied between 25 cents and 1/4 cent per lb. Certain producer associations notably that for cigar tobacco have rejected support prices because of the cost of managing a no net-cost support programme. Initially all of the funding costs were borne by the respective grower associations but in 1986 tobacco manufacturers agreed to finance half of the cost on the basis that there would be higher quotas and that prices would be cut (there was an 18 % cut in support prices in 1986).

Management by the producer associations 6.13. There was a build-up of stocks of tobacco up to 1985 which was dealt with by agreement between the main producer associations and the manufacturers. The manufacturers agreed to purchase the excess stocks from the associations over an eight-year period. This has had the effect of reducing the cost to the growers of running the programme. In addition, in recent years the cooperative for flue cured tobacco has sold 65 Mio pounds of tobacco from their loan stocks at prices discounted by 20 % and 30 %. The aim of this was to reduce the quantities covered by loans and thus the cost of running the programme for the farmers.

Moreover, as stocks are taken into account when calculating quotas it had the effect of avoiding future cuts in quota. 6.14. The US producers, like those in the Community, are under pressure from low priced imports. Thus the manufacturers and some of the producer associations consider that, in future, there may have to be a further lowering of support prices in order to compete with lower priced imports, - an approach that is markedly different from the normal Community method of increasing aid levels under similar circumstances. Another alternative being considered by one producer association is the production of a percentage of the tobacco without guaranteed support prices.

There would thus be two prices for similar US tobaccos, one at the normal price and another unsupported, to compete with imports. 6.15. The pressure from imports has in fact increased due to the purchasing policy of the manufacturing multinationals. In 1990, for example, there were about 100 000 curing barns in Brazil for flue cured and Burley tobacco. One major multinational gave aid to producer groups to increase capacity and in 1991 a further 20 000 new curing barns were built. 6.16. The producer associations enter into storage contracts for the storage of the tobacco that they purchase. The overall storage costs are between 2 and 2,2 cents/lb per year which includes:

(a)interest on the cost of the loan to purchase the tobacco; (b)fumigation; (c)personnel costs for independent supervision; (d)travelling costs to inspect stocks. These figures compare very favourably with the Commission rate of reimbursement to Member States of 0,168 ECU/kg per year (just over 9 cents/lb). While the costs are, of course, not directly comparable, in view of the divergence between them the Commission should re-examine the basis for the calculation of the Community rate. The influence of prices 6.17. Figures provided by the University of Kentucky indicate that the net income per acre from tobacco is 1 000-1 500 USD (2 160-3 240 ECU/ha), whereas the current gross income for soya production is only 250 USD/acre (540 ECU/ha).

Thus, there appears to be a sufficient margin in producer income for price decreases to be absorbed. The Community faces a similar problem. The move towards low price cigarettes may increase the market potential of Community tobaccos, as less of the top tasting and higher priced US, Brazilian and Zimbabwe tobaccos will be used. However, the Community producers will, like those in the US, come under price pressure to decrease market prices to enable them to compete with imports and on the world market. Assuming that the European taxpayer will not continue indefinitely to support tobacco growing at the present level of expenditure, let alone increase aid in order to remain competitive, price cuts must be faced.

In addition, overall efficiency must be improved in order to keep costs down and remain competitive. 7. CONCLUSION 7.1. The Court concludes that the financial situation of the CMO for raw tobacco is not properly presented in the Commission's consolidated accounts in so far as it: (a)does not accurately disclose all relevant information; annual expenditure is not definitive until clearance procedures including any disallowances decided by the Commission have been completed for a particular year; total expenditure on tobacco is either understated in a given year when profits on sales of intervention tobacco are accounted for as negative expenditure, or are overstated as far as the amount of the depreciation leads to under-estimated stock values; total EAGGF expenditure is understated as disallowances are also accounted for as negative expenditure; depreciation of the value of intervention stocks is not based on the actual stock quantities at the date of the consolidated accounts (see paragraph 3.2);

(b)substantial amounts of premiums are either advances which have not yet been cleared as the tobacco has not left supervision or are definitive payments relating to tobacco harvested up to eleven years previously (see paragraph 3.2 (a) and (b)); (c)intervention costs were booked to premiums thus giving the false impression that the related tobacco had been marketed (see paragraph 3.3). 7.2. In general the systems operated by the Commission and those implemented by the Member States, for the control of processing and storage and the taking into and release from supervision gave no assurance that in the three main producing countries expenditure was incurred in a regular and lawful manner (see paragraphs 3.25 - 3.69 and 3.89 - 3.117).

While the Court's audit concentrated on the years 1987 to 1992, the control systems have existed for many years and it is impossible to quantify the full extent of the loss. 7.3. The procedures adopted by the Commission for selling intervention tobacco by tender (see paragraphs 4.39 - 4.61) did not permit the Commission to obtain the best possible prices for the produce sold. Furthermore, the Commmission took insufficient precautions to protect the Community's interests and in particular to prevent irregular manipulation of the tender system. The Commission's management of intervention sales was so poor that at one stage it did not know the market value of the tobacco being sold.

Lots were so large that the tenders favoured a few and, being also of mixed varieties, left the Commission open to the operation of a cartel. The fact that minimum selling prices were not always calculated exacerbated the problem. Furthermore, the promulgation by the Greek authorities of detailed minutes of confidential meetings did not help market management (see paragraph 4.62). 7.4. The Commission should carry out a complete review of the control systems in the Member States. All weaknesses noted by the Court should be particularly examined and steps taken to ensure that controls are strengthened, that payments are made only on the basis of the rules laid down in the regulations and that all incorrect payments are recovered.

7.5. Where there are indications of irregularities regarding contracted varieties (see paragraphs 3.19, 3.55 and 3.102 et seq.) the Commission should ensure that the matter is fully followed up by the Member States concerned. The failure to apply the 'Commune rule` in Greece and Spain (see paragraphs 3.29 and 3.50) should similarly be examined and premiums paid in error recovered. 7.6. The Commission should also examine the evidence of export and arrival in third countries for all exportations of tobacco from Greece from 1989 onwards in order to ensure that payments were regularly made. Where export refunds were paid without satisfactory evidence recoveries should be required (see paragraph 3.138).

7.7. When carrying out the clearance of accounts the Commission should pay particular attention to low value exports to ensure that they comply with the conditions for granting export refunds (see paragraph 3.141); furthermore the Commission should ensure that Member States' controls are strengthened so that, in future, consignments should be physically checked at the point of export to ensure that they are of fair, sound and merchantable quality and such examination should be based on risk analysis, including value (see paragraph 3.141). The irregular basis used for making export payments in Spain should similarly be addressed (see paragraph 3.139 et seq.).

7.8. While the levy charged to partly finance the Greek Tobacco Board EOK. (see paragraph 3.27) is compatible with the concept of the payment of a minimum price for raw tobacco, the Commission should examine whether the charge on Greek processors and in consequence on the growers themselves is compatible with the CMO and in particular Article 40 (3) of the EEC Treaty. 7.9. The Court found that the Commission's central management was unsatisfactory. Information was not properly gathered, or used (see paragraphs 4.3 - 4.11) and world market prices, processing costs and losses and producer incomes were either not properly monitored or were not correctly calculated (see paragraphs 4.12 - 4.32).

Given that the CMO was based on a deficiency payment system the failure to obtain details of the costs of production and world market prices was most unsatisfactory. 7.10. The Commission's management of two of the main policies of the CMO: to convert to varieties in demand and to promote quality, has largely been a failure. Substantial sums of Community expenditure have been wasted on unwanted production and the wrong signals have been given to growers. Despite some recent progress, the reform programme has been set back by a number of years, (see paragraphs 4.63 - 4.77 and 4.81 - 4.83). The failure to lower premium rates when MGQ's were exceeded largely negated the effectiveness of the application of MGQ's (see paragraphs 4.78 - 4.80).

7.11. The source of many of the control problems in the Member States was that payments were made to farmers by processors for unprocessed leaf tobacco, but the ultimate right to the premium financing these payments was not acquired until the processed tobacco had left supervision. This resulted in substantial additional budgetary cost, (see paragraphs 3.4 - 3.6, 3.34 - 3.36 and 3.84 - 3.85). Furthermore, as the lots of leaf tobacco delivered by the farmers lose their identity during processing, the whole process from delivery of raw tobacco to sale of processed bales had to be kept under permanent control in order to assure the physical link between tobacco delivered and tobacco sold.

This was not only administratively difficult and costly for both the administration and the processors but was vulnerable to irregularities. 7.12. The merits of the reform of the CMO consist in the short term of an apparent simplification of the management of the CMO by abolishing price fixing and reducing the number of different premiums. In the long term the foreseen abolition of intervention and export refunds may reduce the possibilities of fraud. However: (a)the extension of the CMO's objectives to those already established under Article 39 of the EEC Treaty increases the need for an improved management information system; the Court found that under the old system, which is heavily dependent on the Member States, the Commission was unable to gather sufficient and relevant information to monitor the achievement of objectives and render transparent the process of premium fixing (see paragraph 4.4 et seq.);

(b)the check that farmers received the minimum prices is replaced by a check that farmers receive the entire premium (see paragraph 2.19 (c)); (c)premiums will only be paid for tobacco of a fixed minimum quality, so moisture, impurities and low grade tobacco must continue to be controlled, some elements of which were not properly controlled under the old system (see paragraph 2.19 (d); (d)reimbursement of the premiums is dependent on verifying that the tobacco has been both processed and marketed thus the process from delivery onwards will still need continuous control (see paragraph 3.8 and 3.16);

(e)as cultivation contracts will remain the legal basis for the payment of premiums, it is essential that all contractual data are independently verified and that actual quantities delivered are checked against contract details (see paragraph 3.10); (f)the distribution and respect of quotas has to be controlled (see paragraph 2.19 (f). 7.13. The new CMO foresees the establishment of expensive control agencies (see paragraph 2.19 (i)) and the creation of additional inter-branch organisations to manage the system (see paragraph 2.19 (j)). In its Annual Report on the financial year 1991 () the Court described the weaknesses in controlling the producer organisations for olive oil in the same Member States as produce most Community tobacco.

Unless the Court's observations are taken into account when allocating responsibilities to the inter-branch organizations, the current control problems identified in this report will not be overcome. The Court considers that the USA producer cooperatives provide a sound example of the way in which such responsibilities should be developed (see paragraph 6.13 et seq.). 7.14. Furthermore, the method of setting the annual quotas in the USA, by taking into account the existing stocks, projected exports and internal demand for the tobacco varieties, provides a relatively accurate method of setting demand and enabling production to be brought into line.

The Commission should examine whether the method can be adopted in the Community in order to set quotas more effectively. 7.15. The fact that quotas may be traded contributes to the creation of economic and efficient farming units in the USA. Although these considerations are also relevant to the Community situation, quotas are not transferable under the Community's market reform. Consideration should be given to whether it would help efficiency in the tobacco market if quotas were eligible to be leased and sold, within clearly defined limits (see paragraph 6.5 et seq.). 7.16. Serious consideration has to be given by the Commission and the Budget Authority as to whether there is adequate justification to support a product which requires massive Community aid and substantial Member State resources to manage and control, when its use results in an enormous social health cost.

Particularly relevant is the fact that half of the Community tobacco production is of such low quality that it is not suitable for the manufacture of tobacco products in the Community and is dumped on third countries' markets using export refunds over and above the premium payments. Thus the Community finances a product for which there is no real demand, instead of urgently seeking a switch to other products in areas where quality tobacco is not, or indeed, cannot be grown. All of these facts combine to show that financing tobacco production is a misuse of public funds. 7.17. It can be anticipated that tobacco production in the Community will come under increasing pressure from imported tobaccos, as is the case in the USA.

Thus, if present policies are pursued, the cost of market support will increase. The direct financing of tobacco production should therefore be reviewed and consideration should be given to the integration of tobacco farmers into the direct income support policy applied to other major markets and into the structural aid system. This would be the appropriate approach to the reform of the tobacco market and would save substantial sums of Community money in the region of 520 Mio ECU per year (see paragraph 4.91 et seq.). It appears that it should be possible broadly to maintain grower incomes through such an alternative approach, while saving 40 % of the present costs of market support.

This report was adopted by the Court of Auditors in Luxembourg at the Court meeting of 21 December 1993. For the Court of Auditors André J. MIDDELHOEK President () Council Regulation (EEC) No 727/70 of 21 April 1970. OJ L 94 of 28.4.1970, p. 1. () Unit of account, preceding the European Unit of account (EUA) and the present ECU. () Special Report of the Court of Auditors No 3/87. OJ C 297 of 6.11.1987, p. 1. () BAT Industries, Philip Morris, RJ Reynolds, Rothmans International. () International tobacco growers' association, 'Tobacco Farming: Sustainable alternatives?` East Grinstead, West Sussex, England, page 1.

() The Economist Intelligence Unit, Tobacco to 1995, Special Report No 2091, London May 1991, page 1. () R.O.W. Sciences Inc.: 'Resisting tobacco in developing countries`. Working papers in support of the 8th World conference on tobacco or health: 'Building a tobacco-free world`. Buenos Aires 1992. () See preamble of Council Regulation (EEC) No 727/70 (see footnote (1)). () Council Regulation (EEC) No 2075/92. OJ L 215 of 30.7.1992, p. 70. () Inter-branch organisations comprise representatives of production, processing and marketing. () Commission of European Communities, D.G. VI-G-1Doc VI/275/92 of 18 May 1993.

() Commission Regulation (EEC) No 2822/88 of 12 September 1988. OJ L 254 of 14.9.1988, p. 6. () Council Regulation (EEC) No 3247/81 of 9 November 1981. OJ L 327 of 14.11.1981, p. 1. () Commission Regulation (EEC) No 1726/70 of 25 August 1970. OJ 191 of 27.8.1970, p. 1. () Council Regulation (EEC) No 1329/90 of 14 May 1990. OJ L 132 of 23.5.1990, p. 25. () Required with effect from the 1992 harvest. () Council Regulation (EEC) No 727/70 of 21 April 1970. OJ L 94 of 28.4.1970, p. 1. () Commission Regulations (EEC) No 1413/91, OJ L 135 of 30.5.1991, p. 15, and No 2376/91. OJ L 217 of 6.8.1991, p. 18.

() Commission Regulation (EEC) No 409/76 of 23 February 1976. OJ L 50 of 26.2.1976, p. 8. () Commission Regulation (EEC) No 1727/70 of 25 August 1970. OJ L 191 of 27.8.1970, p. 5. () Council Regulation (EEC) No 729/70 of 21 April 1970. OJ L 94 of 28.4.1970, p. 13. () Council Regulation (EEC) No 4045/89 of 21 December 1989. OJ L 388 of 30.12.1989, p. 18. () Council Regulation (EEC) No 729/70 of 21 April 1970. OJ L 94 of 28.4.1970, p. 13. () Council Regulation (EEC) No 727/70 of 21 April 1970. OJ L 94 of 28.4.1970, p. 1. () Council Regulation (EEC) No 4045/89 of 21 December 1989. OJ L 388 of 30.12.1989, p.

18. () Commission Regulation (EEC) No 1726/70 of 25 August 1970. OJ 191 of 27.8.1970, p. 1. () Required with effect from the 1992 harvest. () Council Regulation (EEC) No 4045/89 of 21 December 1989. OJ L 388 of 30.12.1989, p. 18. () Commission Regulation (EEC) No 1726/70 of 25 August 1970. OJ 191 of 27.8.1970, p. 1. () Article 10 of Greek national legislative decree 3758/57 combined with Article 12 § 1 of Executive Order Law 1054/46 and Article 5 § 1 of Executive Order Law 1059/38. () Council Regulation (EEC) No 4045/89 of 21 December 1989. OJ L 388 of 30.12.1989, p. 18. () Council Regulation (EEC) No 727/70 of 21 April 1970.

OJ L 94 of 28.4.1970, p. 1. () Council Regulation (EEC) No 2268/88 of 19 July 1988. OJ L 199 of 26.7.1988, p. 20. () Council Regulation (EEC) No 1329/90 of 14 May 1990. OJ L 132 of 23.5.1990. () Council Regulation (EEC) No 1467/70 of 20 July 1970. OJ L 164 of 27.7.1970, p. 32. () Commission Regulation (EEC) No 3665/87 of 27 November 1987. OJ L 351 of 14.12.1987. () Commission Regulation (EEC) No 137/90 of 4 January 1990. OJ L 16 of 20.1.1990, p. 9. () Source: Eurostat. () Council Regulation (EEC) No 727/70 of 21 April 1970. OJ L 94 of 28.4.1970, p. 1. () Commission Regulation (EEC) No 1076/78 of 23 May 1978.

OJ L 136 of 24.5.78, p. 8. () DG VI/E4/ TABAC/3134 of 15 January 1990. () Council Regulation (EEC) No 727/70 of 21 April 1970. OJ L 94 of 28.4.1970, p. 1. () Commission Regulation (EEC) No 1076/78 of 23 May 1978. OJ L 136 of 24.5.78, p. 8. () Special Report of the Court of Auditors No 3/87. OJ C 297 of 6.11.1987, p. 1. () Commission, DG VI-E-3, VI 1087/85 of 23.1.1985. () Special Report of the Court of Auditors No 3/87. OJ C 297 of 6.11.1987, p. 1. () Council Regulation (EEC) No 2075/92. OJ L 215 of 30.7.1992, p. 70. () 'Groupe du Tabac` Report dated 14.7.1990. () Article 4 of Council Regulation (EEC) No 2075/92 (see footnote (9)).

() Commission Regulation (EEC) No 1726/70 of 25 August 1970. OJ 191 of 27.8.1970, p. 1. () Council Regulation (EEC) No 727/70 of 21 April 1970. OJ L 94 of 28.4.1970, p. 1. () Commission Regulation (EEC) No 3389/73 of 13 December 1973. OJ L 345 of 15.12.1973, p. 47. () Commission Regulation (EEC) No 2611/90 of 10 September 1990. OJ L 248 of 11.9.1990, p. 7. () Commission Regulation (EEC) No 3389/73 of 13 December 1973. OJ L 345 of 15.12.1973, p. 47. () Commission Regulation (EEC) No 881/93 of 15 April 1993. OJ L 92 of 16.4.1993, p. 21. () There was no MGQ for 1987, the calculations assume that the MGQ for 1988 also applied to 1987.

() Council Regulation (EEC) No 2076/92 of 30 June 1992, OJ L 215 of 30.7.92, p. 77. () Commission Regulation (EEC) No 3616/92 of 15 December 1992. OJ L 367 of 16.12.1992, p. 13. () Commission Regulation (EEC) No 881/93 of 15 April 1993. OJ L 92 of 16.4.1993, p. 21. () Council Regulation (EEC) No 2075/92. OJ L 215 of 30.7.1992, p. 70. () Council Directive 90/239/EEC of 17 May 1990 on the approximation of the laws, regulations and administrative provisions of the Member States concerning the maximum tar yield of cigarettes. OJ L 137 of 30.5.1990, p. 36. () Decision of the Council and the representatives of the Governments of the Member States meeting within the Council on 17 May 1990 adopting a 1990 to 1994 action plan in the context of the 'Europe against Cancer` programme, Annex, A.

Prevention of tobacco consumption. () Special Report of the Court of Auditors No 3/87. OJ C 297 of 6.11.1987, p. 1. () Commission Regulation (EEC) No 1210/92 of 12 May 1992. OJ L 127 of 13.5.92, p. 5. () Commission Regulation (EEC) No 1076/78 of 23 May 1978. OJ L 136 of 24.5.78, p. 8. () Working Paper of the Commission, DG VI.E.4, of 21.2.92 ST. () The Court of Justice case C 368/89 (Crispoltini) of 11 July 1991. () Commission Regulation (EEC) No 2907/92 of 6 October 1992. OJ L 291 of 7.10.92, p. 6. () Council Regulation (EEC) No 2062/92 of 30 June 1992, OJ L 215 of 30.7.92, p. 22. () Council Regulations fixing the MGQ's; latest:

Council Regulation (EEC) No 861/92 of 30 March 1992, OJ L 91 of 7.4.92, p. 2. Commission Regulations determining production; latest: Commission Regulation (EEC) No 2065/93 of 27 July 1993, OJ L 187 of 29.7.93, p. 26. () Special Report of the Court of Auditors No 3/87. OJ C 297 of 6.11.1987, p. 1. () PIEDA, The Tobacco Industry in the European Community, 1990. Study of January 1992. Total income in growing and processing is estimated to be 744 Mio ECU or 4200 per full-time equivalent job (FTE). With a deduction of 76 Mio ECU for the 18250 FTE workers in processing there is 667 Mio ECU in income in growing.

With 211662 ha. in production in 1990 the average income is 3151 ECU/ha. () Figures for aid by Member State are taken from Table 4 and on area from the Commission's publication, The Agricultural Situation in the Community. () Figures on aid and areas cultivated for the different crops are from the Commission, The Agriculture Situation in the Community, 1992 Report. () Commission, FADN, Farm Incomes in the European Community 1990/91 including selected results for 1986/87 to 1989/90. () Commission DG VI A3. Note of 8.8.1991, ref. 022713. () OJ C 144 of 3.6.1991, p. 23. () WHO Regional Office for Europe, 'A 5 Year Action Plan`, 1988;

Table 1. () Two studies: Dr. A. Sasco and Dr. L. Tomatis, ' Prevention of cancer: prospects for the near future`, Social Europe, 1991 No. 1, p. 10-14. Marion Merrell Dow Inc., 'The Economic Impact of Smoking`. Published by Medical Information Services Inc. 1991. () Marion Merrell Dow Inc., the series 'The Economic Impact of Smoking`, Published by Medical Information Services Inc., 1991. () Koop, E., 1986, 'Tobacco or Health: Talking Points`, Speech given by Surgeon General of the USA at the 39th World Health Assembly, agenda item 22. () Office of Technology Assessment of the US Congress. 'Smoking-related deaths and financial costs`.

September 1985, p. 50. () The Advocacy Institute for the American Cancer Society, 'What Tobacco Really Does To Your Country's Economic Health`. Paper presented at the 8th World Conference on Tobacco and Health in Buenos Aires, 30 March - 3 April 1992. () Council action plans: OJ C 184 of 23.7.1986. OJ L 137 of 30.5.1990. () 'Europe against Cancer. Evaluation of Anti-smoking`, May 1991. () Bureau for Action on Smoking Prevention: BASP Newsletter No 12, Brussels April 1991. () Markku Pekurinen 'Economic Aspects of Smoking`, National Agency for Welfare and Health, Finland, Research Reports 16/1991.

() Council Directive (EEC) No 92/79 of 19 October 1992. OJ L 316 of 31.10.1992, p. 8. () Council Directive (EEC) No 92/80 of 19 October 1992. OJ L 316 of 31.10.1992, p. 10. () PIEDA, The Tobacco Industry in the European Community, 1990. Study of January 1992. () Lewit, E.M., and Coate, D. 'The potential for using excise taxes to reduce smoking`, Journal of Health Economics 1982; 1(2): 121-145. () Lewit, E.M., Coate, D., and Grossman, M. 'The effects of government regulation on teenage smoking`. Journal of Law and Economics 1981; 24: 545-569. () World Development Report 1993. Investing in Health World Development Indicators p.

89, Box 4.3. () In order to ensure that the manufacturers' purchasing intentions are correctly stated, they face a penalty if they buy less than 90 % of their estimate. The penalty is levied on the basis of each pound not purchased and is twice the no-net cost assessment, i.e. twice the cost per pound of running the programme. () OJ C 330 of 15.12.1992, p. 333. COMMISSION REPLY The Commission would make a number of introductory comments on the Court of Auditors' report on the common organization of the market in tobacco. (i) In the Commission's opinion the disclosure of unacceptable behaviour should not cast a shadow on the objectives and general operation of the former market organization; nor should the new market organization, which was specifically designed to remedy past shortcomings, be judged in the light of this bad record.

(ii) Secondly, the Commission would underline that many of the Court's findings have already been mentioned in various Commission inspection reports or in inquiries it has carried out. It would also point out that, following memos from its departments, it has corrected these shortcomings during the clearance of accounts or in the market decisions which it has taken. (iii) Finally, the Commission would point out that the Council also bears responsibility for the content of the market organization, as do the Member States for the administration, management and control of the rules governing the market organization.

Staff increases in the Commission's inspection departments will boost their effectiveness and also the smooth operation of the new market organization. 1. INTRODUCTION 1.1. Tobacco is a demanding and labour-intensive crop. Depending on the variety grown, a hectare of tobacco requires an average of 1 500 hours of labour (2 150 hours per hectare in the case of the oriental varieties in Greece), compared with some 10 hours per hectare for cereals. The production aid for tobacco is therefore, to some extent, aid for regional employment. The Court recognizes at 2.2. that 'tobacco growing is labour-intensive, requiring careful soil preparation, weeding and, for some varieties, intensive irrigation`.

For many farming families, their earnings from tobacco are therefore their only annual income. For the Community, this is the whole point of the measures to help 'rural society`. 1.2. The support granted to Community tobacco-producers is not in contradiction with the Commission's efforts to reduce tobacco consumption under the 'Europe against cancer` campaign. The Community imports more than 70% of the tobacco required for the manufacture of cigarettes, cigars, etc. The tobacco consumed is therefore obtained as much from imports as from domestic production and these imports have by no means increased as a result of premiums for Community producers.

Tobacco would, of course, still be consumed even if the Community withdrew its support for tobacco-growers, thus ending tobacco production in Europe; the effect on public health would be zero. Having said that, it should be pointed out that the Community fund for research and information is financed by deductions from the premiums granted to tobacco-producers (Regulation (EEC) No 2427/93 of 1 September 1993, OJ L 223 of 2.9.1993, p. 3). This 'tax` is to be used to finance programmes of research and information to promote greater knowledge of the harmful effects of tobacco in all its forms, especially among the young.

1.5. The report referred to in this paragraph concerned the alleged behaviour of a Commission official. This official had been relieved of any responsibility for the tobacco market but was the subject of an internal enquiry for which the report was relevant. As soon as the enquiry was ended by the death of the official concerned Commissioner Steichen wrote to Mr Androutsopoulos on 23 April 1993 agreeing that the report could be examined by the Court of Auditors. Mr Androutsopoulos replied on 29 April agreeing to this. On the same day the Secretary-General replied to Mr Androutsopoulos indicating where the report could be examined by the Court.

This is the first time that the Commission was aware that there was any dissatisfaction on the Court's part about the arrangements made in April for studying the report. On the question of alleged fraud, the Commission also took immediate action after the death of the official brought the internal enquiry to an end. The Secretary-General of the Commission wrote on 2 April to the Italian Greek authorities informing them of the allegations and inviting them to take the necessary action on the basis of Article 6 of Regulation (EEC) No 595/91 of 4 March 1991. At the same time, the Secretary-General wrote to the Belgian authorities, in case there were need at any time for cooperation between the Italian, Greek and Belgian authorities.

On 16 July the Commission wrote again to the Italian and Belgian authorities drawing attention to the possibility that irregularities had been committed which might justify the opening of judicial proceedings on the recovery of funds in favour of the Community. With these letters the Commission sent copies of a summary of all the tobacco sales between 1986 and 1990 with details of the successful tenderers. The Commission also indicated its availability for a meeting to explain the elements which could be helpful to those responsible for the enquiries in these Member States. Copies of all these letters and documents are available to the Court of Auditors.

1.6. The letter of assignment which the Director-General for Agriculture sent to one of his former colleagues, instructing him to investigate the operations of the Tobacco Division, is contained in the annex. 1.8. The Commission has asked the Member States to take such action whenever it has been able to do so, i.e. when it has possessed sufficient evidence. 1.9. The purpose of the new rules introduced in 1992 was to put an end to the manfunctioning encountered in the past. Financial corrections have been imposed on the Member States for mismanagement and inadequate controls under the old market organization.

2. THE MARKET FOR RAW TOBACCO The EC market The Common Market Organization (CMO) 2.19. It was after detecting abuses in the application of the rules in late 1989 that the Commission decided to conduct a major overhaul of the market organization for tobacco and instructed a group of independent experts to consider what form the market organization should take in future. 3. EXAMINATION OF THE ACCOUNTS AND OF THE UNDERLYING TRANSACTIONS IN THE MEMBER STATES 3.2. (a) As in the case with the other market organizations, disposal of a crop and the corresponding expenditure may be carried over to the following financial year.

This phenomenon is undeniably more common in the tobacco sector than elsewhere, but this is because the product must be dried and then repeatedly processed before being sold. The premiums for one year's harvest are therefore paid mainly from the following year's budget. - Although the budgetary nomenclature makes no distinction as far as the payment of tobacco premiums is concerned, the Commission requires the Member States, in the interests of control and sound management, to break down annual payments by date of harvest. (b) When drawing up the budget, provision must be made for any expenditure which might be incurred, irrespective of whether it takes the form of an advance or a balance.

Obviously, this breakdown cannot be known in advance. Each budget therefore covers both the advances for one crop and the tobacco premiums outstanding from previous crops, payable on completion of the control. The expenditure declared is not final until adoption of the decision on the clearance of accounts for the year during which the payments took place. Payments in the form of an advance are not final until the eligibility and regularity of the expenditure can be checked and this is normally done when the accounts are cleared. - On the other hand, the payments made by Greece in 1992 in response to a request by the Commission during an inspection visit mainly took the form of the premiums outstanding on tobacco placed under supervision between 1981 and 1986.

In order to identify these specific payments and check when the balances of the said premiums were paid, the budgetary authority entered ECU 17 million under the heading 'Other` in supplementary and amending budget No 3/92 (the Commission takes this entry into consideration at market management level). (c) The depreciation of intervention stocks is governed by Article 8 of Council Regulation (EEC) No 1883/78 laying down general rules for the financing of interventions by the EAGGF Guarantee Section. The Commission already replied to the Court on 5 April 1993 on the method of calculation used. This provision is linked not to the Community value of the tobacco stored but to the foreseeable sales value, which is always based on the average prices of previous tenders.

The aim is not to pass on to future financial years the potential losses on the tobacco bought in. In view of the fairly low sales prices obtained in previous years, it is quite logical that the Commission took these as a basis for determining the probable prices for 1991 and 1992. The Commission reduced the number of tenders for the exportation of intervention tobacco for a number of reasons, including the difficulty of obtaining the real price on the world market. This approach paid off (as the Court itself recognizes in Chapter 3) since selling prices subsequently rose far higher than they had been in the past.

The Commission therefore made a profit on sales. It should also be pointed out that: - The EAGGF achieved considerable savings in financing costs over the whole storage period (the capital needed to purchase intervention products is supplied by the Member States and then refunded by the EAGGF). Refunding part of the value of the goods through depreciation reduces the amount of capital tied up and, consequently, the financing costs involved. - Profits on sales are registered on payment of the goods sold and are immediately credited to the EAGGF. (d) This is the normal clearance procedure. (e) Profits on sales in a given financial year are entered in the accounts as negative expenditure for the year itself.

The sales operations form part of the market management measures and it is therefore legitimate that their financial implications should be entered in the accounts with the other expenditure. This is true for all products covered by the market organizations and not just tobacco. The profits on sales are not mentioned since these amounts are not to be considered as advances but form an integral part of the EAGGF's storage operations in a given financial year. - The amounts recovered during the clearance of accounts are entered in the appropriate chapter of the budget (B1-37) for the financial year during which recovery took place, in accordance with the Financial Regulation.

These operations are in no way treated as a carryover from one financial year to the next. 3.3. The premiums were introduced to guarantee planters a minimum price. The planter is entitled to receive this minimum price irrespective of whether the tobacco is sold to the first-stage processor or to the intervention agency. The price paid on intervention therefore consists of two components: - the premium and - the value of the tobacco at a price level comparable to that paid by the processor (effectively the Community market price). If the premium is included in the value as the Court requests, either depreciation would need to be considerably higher in order to align the value of the stock to the foreseeable sales price or higher potential losses would be carried over to future financial years, the cost of financing would rise and the capital provided by the Member States would have to be increased.

3.4 3.6. The Commission will adopt forthwith the appropriate provisions to limit expenditure due to the agricultural conversion rates. Premiums 3.7. The premiums are paid in the form of an advance (see 3.8). 3.16. The Commission agrees with the Court. That is why the only means of comparison is a quantity control by variety or group. Germany 3.19. The link between the premium certificates and the contracts concluded in Italy between German operators and Italian producers was brought up in an official letter (No 52246 of 9 February 1990) which the EAGGF addressed to the German authorities during clearance of the 1988 accounts.

The Commission has examined the German authorities' reply. 3.20. Article 5 of Regulation (EEC) No 1726/70 allows the German authorities to act in this way. 3.22 and 3.23. The problem of paying the producer the minimum price before granting the premium was dealt with in a specific point (3.9.1.2.) of the summary report on the clearance of the accounts for 1984 and 1985. The same point was raised in a reminder sent to the German authorities in 1990 in connection with the clearance of the 1988 accounts. In their reply the German authorities undertook to act in accordance with Community rules. 3.24.

The stock levels of the 21 German processors do not constitute a major problem in view of the quantities qualifying for a premium (around 6 500 tonnes) and the outlets which exist for the varieties granted a premium. Greece 3.28. The Court notes that there is no land register in the tobacco-growing areas of Greece. Since tobacco is a rotational crop, no cultivation contract was needed under Community rules until the 1989 harvest and this Greek tobacco could even be processed in another Member State. 3.29. Under the Greek structural programme (1986 IMP and 1990 MOP) there was provision for transferring production towards Virginia.

The list of eligible communes did not therefore require adjustment. Furthermore, the authorization given by the Council not to demand contracts made it more difficult to control the eligibility of communes up to 1990. 3.30. The Commission has already planned a specific inquiry into the volume of tobacco receiving a premium which was produced in ineligible communes from the 1989 and subsequent harvests and paid for during 1991. Contacts have been established with the Greek authorities to ascertain the amount of the correction. 3.31 and 3.32. The EOK does indeed handle certificates in a discriminatory fashion to make up for the lack of genuine contracts between processors and producers.

It 'stocks` the certificates in its drawers and issues them one at a time whenever necessary. In all probability, a financial correction was to be made in the clearance of the 1991 accounts since the Community rules had not been respected. Spain 3.42 to 3.44. The Commission is aware of the problem raised by the Court since the Court was notified of it by a Commission memo in December 1990 and it is referred to in the report on an inspection carried out from 27 to 31 January 1992. The question was brought up in an official letter sent to the Spanish authorities in February 1992. Since then the case has been dealt with by the appropriate Commission departments.

3.45 to 3.48. The Commission has long been aware of the question of duplicate contracts. The Spanish authorities' controls have been complicated by the fact that the Spanish land register has not been kept up-to-date. This shortcoming has, however, been remedied. Since the 1991 harvest, the Council has made the processing premium conditional on the establishment of a single cultivation contract with the producers (Regulation (EEC) No 1329/90). Since the 1992 harvest, Regulation (EEC) No 1197/92 has also stipulated that at least 5% per variety or group of varieties of the cultivation contracts and declarations must be checked in the case of each processing enterprise.

3.49. The Court has been informed of the Commission's contracts with the Spanish authorities. 3.51. The question raised was dealt with in the official letter addressed to the Spanish authorities in January 1991 in connection with the clearance of the 1989 accounts. The Spanish authorities undertook to alter their system for drawing up the premium certificate. 3.53. et seq. In connection with the clearance of the 1990 accounts, an inspection visit took place in January 1992 to clarify the question of Burley E and Burley F. In the opinion of several tobacco experts, Burley E is genetically and botanically a Burley tobacco.

Any difference results from the conditions of cultivation, the micro-climate or the soil. Nevertheless, all Burley can be processed. The difference in the premium for the two types of Burley is due to the fact that processing costs far more than fermentation. In March 1992 the EAGGF informed the Spanish authorities in an official letter that, under Community rules, no more than one cultivation contract for the same plot of land could be concluded with tobacco processors and that stricter checks should be carried out on point 1 of the Annex to Regulation (EEC) No 1726/70. 3.58. The ANT in Spain did indeed decide to suspend the payment of premiums pending checks.

Where these checks led to the detection of fraud, repayment of the premium was demanded. As a result of these controls, the quantity of tobacco claimed to be Burley E dropped sharply in subsequent financial years, as the statistics show. 3.60. Since the 1993 harvest, the premium has been the same for Burley E (light air-cured) as for Burley F (dark air-cured). 3.61 to 3.63. The Commission is aware of this question and has tried to resolve the problem by proposing a 3% reduction in the maximum moisture content and no longer allowing the increase up to this maximum. 3.64. The Commission would draw the Court's attention to the fact that the procedure for determining the moisture content is practical in nature, that it should be treated with caution and that the results should take the form of bandwidths.

Averages should be used since two analyses of the same batch will never give the same result. In January 1991 the Commission wrote to the Spanish authorities asking them to comply with Community rules by making the SENPA inspectors carry out their real role (i.e. help determine tobacco quality) and setting up an appeal board in the event of any disagreement on the determination of tobacco quality and the consequences as regards the minimum price to be paid to the producer. 3.65. The Court notes the similarity between the samples of Burley E and Burley F. This is quite normal as long as the Burley F has not been fermented.

3.66. The question of the tobacco destroyed by fire in 1989 and 1990 was resolved during clearance of the 1990 accounts when the Spanish authorities decided to make payment in advance of the clearance decision. 3.67. Specific or technical problems will always arise with inspections. The Commission would point out that the inspections in Spain are carried out without prior notice and are therefore very effective and that there is an inspectorate with adequate staff to carry out regular inspections in the tobacco sector. France 3.70 to 3.84. The Commission analysed the system of tobacco management and inspection in France during clearance of the accounts for 1988.

In a letter dated February 1990 the EAGGF addressed to the French authorities the same comments as those made by the Court of Auditors in its report. The French authorities subsequently altered a number of procedures and considerably stepped up their controls. 3.76. The Court's comment applies to all Member States. Having said this, the Community moisture content is a maximum; in practice, the premiums are always paid on the basis of this maximum. 3.79. In France the Commission has evidence that the purchase price is passed on to producers. The question of agricultural conversion rates in the tobacco sector is still being discussed.

Italy 3.91 and 3.92. Regulation (EEC) No 1726/70 requires purchasers and persons filing cultivation declarations to have their cultivation contracts and declarations registered by 1 August at the latest. Since the 1990 harvest, the Ministerial Decree of 5 April 1990 obliges processors to submit their cultivation contracts and/or declarations (with signatures authenticated) before 20 June, together with a diskette enabling cross-checks to be carried out. 3.93 to 3.96. Until the 1992 harvest, the Community rules called for controls when leaf tobacco was placed under supervision and when tobacco which had undergone first-stage processing and preparation left the place of supervision.

Italy introduced supplementary measures in its National Decree of 31 March 1979: random checks of areas under contract, a delivery timetable for each variety, and checks during the processing of leaf tobacco. National instructions require registers of leaf tobacco, inventories of leaf tobacco placed under supervision, and registers of processed tobacco. From the 1992 crop, Community rules stipulate that at least 5% of the cultivation contracts and declarations registered per variety or group of varieties must be checked in the case of each processing enterprise. 3.97. The Commission would point out that, in its letters of 11 January 1991 and 24 February 1992, it underlined the obligation to develop systematic control arrangements when tobacco is placed under supervision and stressed the role to be played by national controllers in determining tobacco quality.

On 18 November 1991 the AIMA decided to make the SITRIS responsible for all controls in the tobacco sector to make up for the shortage of staff. 3.98. In the inspections carried out by the EAGGF, especially in the case of producers, its officials have always insisted on checking whether the minimum price has actually been paid to the producers, either by questioning those concerned about their declarations or by asking for bank statements or proof of payment. 3.99. Once the tobacco has been supplied by the producers, it is stored by the processing enterprise in a way which reflects the commercial category preferred by manufacturers or desired by future purchasers.

3.100. It is not true to say that there were no checks. Each premium file is closed by conducting a final comparison to ascertain, in the case of each premium certificate, that the tobacco which has entered has also left. 3.101. Once the tobacco has been processed in accordance with commercial requirements, the processed tobacco bales cannot be traced to a specific planter. However, the processed bales bear a number and show the year of the harvest, the variety, the quality and the position on the stalk. Stock movements are recorded in the registers kept by each store to show entries and withdrawals of baled tobacco.

3.102. The Commission recognizes that it should have reacted sooner to the production changes which were taking shape. It could have taken appropriate measures under Article 13 of Regulation (EEC) No 727/70. The reluctance of lawyers to accept corrective measures in the course of a financial year has been backed up by the Court of Justice which, in its judgment on the Crispolini case in July 1991, found against the Council and the Commission for changing the rules in the course of the marketing year. 3.103 and 3.114. The Commission conducted a special investigation into Badischer Geudertheimer in June 1989.

The increase in the areas planted with Badischer Geudertheimer and the quantities produced was to the detriment of the Havanna and Burley varieties. This rapid expansion of Badischer Geudertheimer was due to the difference in premiums between the varieties concerned (25%). Since then it has been found that mainly Badischer Geudertheimer hybrids were involved. That is why Annex III of Regulation (EEC) No 1252/89 introduced different maximum guaranteed quantities for area A (Germany and France) and area B (Italy) from the 1989 harvest onwards. Different prices and premiums were also introduced for each area from the 1991 harvest onwards.

Under Regulation (EEC) No 2436/91 the Commission offered for tender and sold from intervention stocks a total of 7 517 715 kg of Badischer Geudertheimer from the 1988 harvest. By the end of 1993 there was no longer any tobacco of this variety in intervention storage from the 1988, 1989, 1990 and 1991 harvests. The level of private stocks of this variety held by processing enterprises is still not known. 3.113. There might be a resemblance. Varieties were not produced for what they were but for what they would fetch. They were often hybrids known for their differences in yield, which explains the differences noted with German or French Badischer Geudertheimer.

3.115 to3.117. The Commission has been closely monitoring the Paraguay variety and the possible replacement of the Badischer Geudertheimer variety by Burley. An inspection was carried out in September 1990. Intervention storage3.122 to 3.125. The problem raised by the Court relates to the Assuntori in Italy. 3.124 and 3.125. This occurred solely because of financial considerations relating to currency differences and the absence of monetary compensatory amounts. Intervention documentation3.126 to 3.129. During the investigations into the quality of tobacco in intervention, which were carried out in February 1988 and January 1990 at all the storage centres holding tobacco sold by tender, the EAGGF officials came into the possession of official documents assessing the quantities placed in intervention.

These documents are entitled 'riepilogo dei risultati della perizia`. The examination covers not every carton but only 25% of those presented for intervention. The assessment of these 25% is then extrapolated to the total volume. It is this assessment which the Commission took as a basis for offering the intervention tobacco for sale. Security of the tobacco in intervention 3.130. The cartons inspected during the two visits in February 1988 and January 1990 were all stamped AIMA. Although the EAGGF is unable to conduct an inventory of all the cartons, it is still the overall quantities which are checked.

The Italian total for 1988 was thus corrected by LIT 2 889 304 358 and its total for 1992 by LIT 9 billion in respect of quantities exceeding the maximum limit on losses. 3.132 to 3.133. A financial correction of DRA 449 423 609 was made during the clearance of the 1990 accounts for the tobacco destroyed by fire in Greece. Export refunds 3.135. The refund may be justified by the fact that customs duty was charged on the tobacco imported into the Community and that its price therefore affects internal market prices. A refund is therefore required to align the Community price to the world price. Under the reform, refunds have been abolished since the 1993 harvest.

Low-value exports 3.136 to 3.141. It is cheaper for the EAGGF to grant refunds for the export of tobacco with a low commercial value than to buy it in for intervention. Indeed, Community rules do not set any limit on intervention purchases of such tobacco, which is completely unsuited to market requirements as regards both quantity and quality. Now that refunds have been abolished under the new market organization any exports which still take place will not entail any costs for the EAGGF. 3.139. The Commission investigates all tobacco exports to certain Eastern European countries. It will pay particular attention to this case.

Weight adjustments in Spain 3.142 to 3.145. The Commission has just received a reply from the Spanish authorities to its letter No 22383 of 30 June 1993 in which it raised the question referred to by the Court of Auditors. The Spanish authorities claim that the Community rules do not clearly specify the criteria to be used in respect of the weight for calculating the refund and that they therefore applied the same criterion as for the calculation of premiums, i.e. the weight as adjusted for moisture. However, Regulation (EEC) No 3665/87, which deals with refunds, specifies that the weight to be applied is the actual net weight.

During clearance of the 1991 accounts the Commission will recover PTA 8 731 111 on the basis of the recalculation of the actual weight. Clearance of accounts 3.146. The Commission will make corrections in respect of cases not settled in the forthcoming clearance procedures for the period covered by the Court's enquiry. 3.147 and 3.148. Since 1988 the Commission has systematically checked the list of outstanding securities during its inspection visits to all the Member States. The situation in Greece as regards premiums for the 1981-86 harvests was thus monitored from the outset. The Court rightly comments that, until the 1990 harvest, the Community rules did not lay down a time limit for the processing and sale of baled tobacco.

However, during clearance of the 1989 accounts, the Commission set a deadline of 31 December 1992 for the harvests in question and recovered the amount indicated by the Court. A financial correction will therefore be made during clearance of the 1991 accounts to recover the premiums for 1 838 168 kg of baled tobacco which was not exported within the time limit laid down. The amount of this correction is not yet known. The adjustment of DR 2 705 095 will be taken into account during this clearance procedure. 3.149. On each inspection visit to a Member State the Commission pays particular attention to the list of outstanding securities.

These lists are monitored year in year out and any delays in the closure of the files are noted. Tables V(a) and V(b) relating to the premiums paid for each financial year make it possible to identify the harvests for which premiums have been paid and then investigate the findings. 3.150 and 3.151. The Commission would assure the Court that the EAGGF departments carefully examined the letter of August 1988; a special enquiry was launched in late 1988 and encountered considerable difficulties in view of the destinations involved (Albania and Bulgaria). It was only after the political changes in these countries that this enquiry could be given a decisive boost.

On the basis of the evidence obtained from its own enquiries the EAGGF made a correction of DR 4 491 969 372 for Greece and LIT 15 697 544 030 for Italy during clearance of the 1990 accounts and reserved the right to examine the regularity of export operations in 1988 and 1989 (investigations are in progress in Greece, Italy, the two destinations mentioned and other countries in the same region). 3.152. (a) The report to which the Court refers was drawn up by two officials responsible for the tobacco sector in 1986. The Court quotes two phrases on page 19 of the report which should not be taken out of context.

The Commission officials spoke of the internal administrative and financial procedures and the inspection arrangements for deliveries of leaf tobacco, processing and withdrawals. The reference to an internal checking system as a solid basis for the organization of the European tobacco market is explained on the same page when it is stated that Spain and Portugal went to Italy to draw inspiration from the Italian system. The Court also forgot an important phrase which precedes the reference to the internal checking system: 'a great level of segregation of duties assuring the necessary checks in different stadia and an internal checking...`.

It is the 'theoretical` procedural system which was considered 'exemplary`. The way it was applied may have been anything but. That is why, on page 23 of the report, the officials responsible for the inspection visit wrote that: 'in order to be able to check the complete flow of documents it is absolutely necessary to go as far as the regional offices`. On page 10 of the report it was also stated that 'in spite of the fact that the common tobacco market organization in Italy covers a great area no other government or private bodies are involved in checking or organizing this market`. The Court can therefore see that major reservations were expressed and that sometimes considerable amounts were recovered when the accounts were cleared.

On page 9 of the report it is also stated that AIMA Tobacco had the following staff at the time of the inspection: - 12 persons for the technical and administrative aspects at headquarters; - 79 inspectors in the eight regional offices. However, when the Court conducted its inspections, a large number of retired staff had not been replaced. In letter No 66999 of 8 December 1986 the Commission therefore asked the Italian authorities to give an explanation of a draft law which would deprive AIMA Tobacco of its staff. At the time the officials conducting the inspection examined sixteen files, all the responsibility of the Perugia regional office.

The Court's comments in paragraphs 3.90 to 3.96 obviously do not refer to these sixteen files, which date from 1986. 3.153. It has to be admitted that the quantities actually produced was never verified since 95% of the tobacco would have been exported when the inspection was carried out. The current enquiry involving Albania and Bulgaria may reveal that much of this tobacco was in fact exported fraudulently. 3.154. The Commission cannot agree with the Court's opinion and would refer to its reply to point 3.152a). As regards the files relating to the Badischer Geudertheimer variety, the Commission would refer to the reply in the report drawn up following the inspection visit of 14-17 June 1989, of which the Court is aware.

3.155. The fact that the AIMA employed a private firm should be seen in connection with the findings which the EAGGF addressed to the Italian authorities in official letters dated 13 February 1990, 11 January 1991 and 24 February 1992 relating to the clearance of the accounts for 1988, 1989 and 1990 respectively. 3.156. The Commission's reply is to be found under the various points relating to Spain and Greece. 4. MANAGEMENT OF THE TOBACCO MARKET 4.1. The Commission approves these management criteria, but they cannot always be met because of the actual market situation - and it is this that the Council and the Commission have to cope with.

The fixing of the basic parameters 4.2. The Commission recognizes that this price mechanism was too complicated and therefore impossible to control. 4.3 to 4.28. The tobacco market is notoriously difficult to monitor. For a number of reasons - the large number of varieties on the Community and world markets, the qualities marketed (with no standardization of characteristics), the limited number of major traders (with a significant role played by the multinationals), rapid fluctuations on the international market and economic policy factors - the fixing of Community prices can never be an essentially mathematical exercise based on objective data.

That is why the Commission, in line with the conclusions of the group of high-level experts, decided to abandon any idea of revising the old market organization and opted instead to replace it by a new and fundamentally simplified system. 4.7. The compilation of statistics is a problem common to all agricultural products. At present, the Commission tries to request the minimum essential. 4.8 and 4.9. Tobacco is not exceptional as compared with several other sectors: when the market organization was set up, national price levels were simply taken over. 4.16. It is not always possible to compare the systems applied in the Community and in non-member countries since the types of tobacco produced in the Community do not lend themselves to the classification adopted by these countries.

4.29 to 4.32. The premiums proposed by the group of experts are set out on page 18 of its report (the Court has a copy). However, the Commission has limited the regional differences in the premiums and incorporated into the total the specific aid amounting to 10% of the premium (e.g. dark air-cured: ECU 1 818 kg + 10% = ECU 2/kg as proposed by the experts). In fixing the premiums, the experts' report took account of both the market demand for the variety and socioeconomic factors. Council Regulation (EEC) No 2075/92 states that 'the purpose of the premium shall be to supplement the incomes of producers whose products correspond to market requirements and to facilitate the disposal of tobacco produced in the Community` (Article 3(3)) and that particular account should be taken 'of past and foreseeable possibilities for the disposal of the various types of tobacco, under normal conditions of competition, on the Community and world markets` (Article 4(1)).

The premiums proposed by the Commission for the 1993 harvest (Council Regulation (EEC) No 2067/92) take these criteria into account. The tobacco market cannot be described as a regulated market since it is open to the world market, given the low level of tariff protection and the absence of quotas or any other technical barriers. 4.32. The Commission cannot allow derogations from rules adopted by the Council without overturning its own management decisions. 4.39 to 4.61. General comment: The Commission would have preferred the Court to deal with this measure by drawing a distinction between the actual facts and the 'irregularities` of management peculiar to this market organization.

Admittedly, there were loopholes in the rules but it was the way in which they were applied that increased the number or extent of these irregularities. At all events, the Commission has taken appropriate measures under the new market organization to avoid any recurrence of these shortcomings and outlined them in a long report sent to Parliament's Committee on Budgetary Control in mid-1993. 4.47 to 4.49. Since the issue of invitations to tender was resumed in September 1991, the procedure has been considerably tightened up to avoid copying errors and, in particular, to avert any risk of irregular operations.

The basic details of the tenders are listed in a number of documents in the presence of several officials, who certify them with their signature. These documents enable cross-checks to be made; eliminating the risk of errors or manipulation. The transcription of tenders (in the form of a telex sent to the Member States without specifying the tenderer) is subject to thorough administrative controls. 4.51. When tenders are evaluated, world market prices are also taken into account, on the basis of official and confidential information at the Commission's disposal. 4.52 to 4.55. The Commission considers it important to remember how difficult it was at that time to establish objectively the market value of different lots consisting of different varieties.

From the information available, the Commission had assessed the first and second lots as being of equal value. Two lots of virtually identical composition were offered for sale during the same tendering procedure - the first lot being the 1 519 836 kg mentioned by the Court and the second lot consisting of 1 519 991 kg. Since being offered for sale by tender on a number of consecutive occasions on 7 November 1988, the first lot rose from DR 17.51/kg to DR 102.00/kg while bids of between DR 18.43/kg and DR 92.00/kg were received for the second lot. Having withdrawn a number of lots, and noting the considerable increase in offers for the lot in question, the Commission intended to show traders that it was ready to resume the selling of lots once prices had increased sufficiently, although it remained cautious about the quantities to be sold.

The Commission also considers it doubtful whether the first lot (1 519 836 kg of Mavra, Kaba, Kulak, Katerini and Burley) should be regarded as much superior in quality to the second lot (425 693 kg of Mavra and Basmas). Basmas generally fetches far higher prices than the tobacco in the first lot and at the time the prices for Mavra were exceedingly low. At all events, the price later obtained for the second lot was far higher than that offered at the time in question. Management of sales 4.57. It should, however, be pointed out that market conditions have developed very favourably for low-quality tobacco as a result of demand from Eastern Europe.

4.58 and 4.59. The number of traders in the baled tobacco sector is limited. At the latest sale by tender, covering eighteen lots of between 122 and 1 073 tonnes on 6 September 1993, the largest lot received the same number of bids as the smallest. The Court's allegations are not borne out by the fact that a reasonable number of tenderers made bids whenever large lots were offered for sale in 1991-92. 4.60 and 4.61. See 1.5 and 1.8. Confidentiality of Management Committee meetings 4.62. It is only too true that Member States give their traders information, including confidential information, on the discussions of the Management Committees.

This is not specific to the tobacco sector. The question also arises of whether, in the interests of transparency, traders directly affected by the Management Committees' decisions should not be informed of the reasons underlying these decisions. Conversions to varieties in demand 4.63. Conversion was bound to fail since tobacco was being grown more for the premiums granted than for its intrinsic quality. 4.65. The rapid increase in the production of Virginia EL in Greece is mainly due to the very high premium since the 1989 harvest and the ineffectiveness of the stabilizers once the maximum guaranteed quantity, which the Commission opposes, has been exceeded (and the overrun is increasing from year to year).

The Commission considers that Greek investment in the flue-cured sector in 1992 was speculative. However, it has implemented a conversion programme to take account of the fact that this investment placed a heavy burden on some small family farms which could not be run profitably under the new quota arrangements. 4.66 and 4.67. The Commission would repeat what it has stated above: The Council knowingly and deliberately implemented a basically social price-support policy; under this approach, the commercial value of the tobacco produced was only secondary. The failure to consider this essential fact when judging EEC policy in the tobacco sector could lead to all sorts of conclusions.

Under the old market organization the Commission nevertheless adopted, or proposed to the Council, a number of measures concerning the varieties posing problems, in particular: (a)a reduction in the maximum guaranteed quantity for the varieties in question: from 25 700 tonnes in 1989 to 21 000 tonnes in 1992 in the case of Geudertheimer hybrids and Forchheimer Havanna II c and from 30 000 tonnes to 26 500 tonnes in the case of Mavra and Tsebelia; (b)a reduction in prices and premiums. Since the 1989 harvest and before application of the stabilizers, the premium has been reduced by 34.6% (ECU 0.876 kg) for Geudertheimer hybrids, 13.1% (ECU 0.251/kg) for Forchheimer Havanna II c, 26.2% (ECU 0.679/kg) for Tsebelia and 26.2% (ECU 0.555/kg) for Mavra.

When the stabilizers are taken into account, the premium actually paid fell by 44.4% for Geudertheimer hybrids and by 37.3% for Tsebelia and Mavra between 1989 and 1991; (c)in the event of the maximum guaranteed quantity being exceeded in 1992, an increase in the cut-off from 15% to 23% (the Commission had proposed 30%); (d)a 30% reduction in premiums for quantities of low-quality tobacco exceeding a certain threshold, from the 1992 harvest onwards; (e)a reduction in the intervention price to 75% of the target price for the varieties causing problems; (f)the establishment of maximum yields to limit the quantities eligible for a premium from the 1992 harvest onwards together with compulsory controls.

The combination of these measures, especially those referred to at c) and d), explains the dramatic fall in the production of black tobacco in Italy (virtually all of inferior quality and thus doubly penalized). The slow but sure decline in the production of these problem varieties should not therefore be underestimated despite the fact that there is no quota arrangement. The Commission's powers are not unlimited. The problems encountered in the tobacco sector are largely due to the fact that the Council and Parliament refused to accept the Commission's proposal that premiums for the maximum guaranteed quantities should be reduced in proportion to the volume produced.

Promotion of quality 4.70 and 4.72. The yields obtained under normal production conditions must be compared with the yields specified in Regulation (EEC) No 2501/87. Normal yields are then found to be: - lower in the case of Round Scafati (No 34), - the same in the case of twelve varieties, including Paraguay (No 4a), Bright (No 7), Burley I (No 8), Forchheimer Havanna II c (No 11a), the Geudertheimer hybrids (No 11c) and Erzegovina (No 15), - slightly higher in the case of the other varieties, with the exception of certain Greek varieties (Tsebelia, Mavra, Virgin EL) showing a more substantial increase.

Finally, this revision should be placed in its context. Regulation (EEC) No 1413/91 having limited the amount of the premium to the maximum yields introduced for the 1992 harvest, the yields specified in Regulation (EEC) No 2501/87 are no longer simply indicative and of agronomic value but determine the amount of aid to which producers are entitled. That is also why both maximum and 'normal` yields have to be laid down. These yields were not based on statistics available to the Commission but were established by the group of experts appointed by the Commission. It should also be pointed out that, contrary to a widespread misunderstanding, an increase in yield is not necessarily synonymous with a reduction in quality.

Any improvement in growing methods usually means an increase in both quality and quantity, up to certain optimum levels. The Commission would also point out that it is the cigarette industry which determines the final characteristics of the products (nicotine, tar, burning quality, flavour, etc.). It is not therefore correct to claim that current demand focuses on the types of cigarette smoked in industrial countries such as the United States and Japan. The countries to the east of the Community are happy with cheap tobacco and the 'problem` varieties find a natural outlet there. 4.73. It is very tricky to compare a US Virginia with an EEC Virginia.

4.74 and 4.75. The Commission was aware of the fact that the Community support granted to some varieties (premiums plus export refunds) acted as an incentive to produce tobacco without any real commercial value. That is why the Commission proposed a fundamental reform of the market organization to the Council and fought hard for the refund to be abolished. 4.76. The purpose of Regulation (EEC) No 1210/92 was not to improve tobacco quality but to refuse payment of the premium for just any sort of tobacco. This brought about a spectacular drop in the production of some problem varieties in 1992. 4.77.

Round Scafati has been dropped from the list of varieties contained in the Annex to Regulation (EEC) No 2075/92. Limitation of production Budgetary stabilizers 4.78. Prices and premiums have always been proposed by the Commission and set by the Council before the stabilizers to be applied to the previous harvest were known. Prices and premiums for the 1991 harvest were cut considerably. The same arrangements were left in place for the 1992 harvest, pending reform of the sector. However, the cut-off was raised from 15% to 23% (the Commission proposed 30% and a reduction of 30% was introduced for poor tobacco varieties about a certain percentage.

In such cases, therefore, the reduction in the premium was no less than 53%. This applied to 70% of the production of certain problem varieties. 4.79. The Court notes that, over the years, the Commission has proposed increasingly higher cut-off figures (from 5% for the 1988 harvest to 23% for the 1992 harvest), reflecting its desire to penalize excessive production. In view of the difficulty of controlling the situation through maximum guaranteed quantities, the Commission proposed, and the Council adopted, the introduction of quotas under the new market organization. 4.80. In view of the rapid increase in expenditure, the Commission, as part of the prices package for 1990/1991, proposed to the Council a number of measures to remedy the situation, including a reduction in maximum guaranteed quantities, target prices, premiums and intervention prices for certain varieties.

As far back as late 1989 the Commission also instructed a group of high-level outside experts to examine the situation in the tobacco sector and to submit proposals for improving the market organization. The group submitted its report in late July 1990. This report serves as a basis for the proposal to reform the market organization submitted by the Commission in 1991 and adopted by the Council in June 1992. The effects of rates of exchange 4.81. The Commission's 1986 proposal to waive the obligation to process tobacco in Greece was dictated by reasons relating to the free movement of goods. However, the absence of MCAs could only lead to speculation.

Farmers' income 4.89 to 4.92. As already stated at 1.1, it is not appropriate to compare the level of support on the basis of area (aid per hectare). The area cultivated is only of secondary importance since it is determined by the availability of the labour needed to farm it. Tobacco is not primarily a product of the soil but a product of the grower's toil. The average area of tobacco farms is small. Withdrawal of Community support for this product would wipe them out. As most of them are small farms giving work to several members of the family and are located in the less-favoured areas of the Community, a measure of this sort would only increase unemployment and social instability in the regions concerned.

Reference to the Community's efforts in the public health sector is not appropriate since Community tobacco could easily be replaced by imported tobacco, as already explained at 1.2. 4.93. In view of the specific nature of the tobacco sector, neither the Commission nor the Council has opted for a completely different approach based on economic criteria alone. The possible reduction in the cost to the Community budget is only one parameter of many. What would be done with the land that became available (the average size of tobacco plantations in the Community is only 1.1 hectares) or with the tobacco producers and all those who earn a living from the cultivation and processing of tobacco.

In short, the only advantage of this alternative proposal would be to make the EEC market wide open... to the multinationals. Cutting production would not reduce tobacco consumption in the Community since demand would be met by imports from non-Community countries. 5. HEALTH RISKS AND OTHER COSTS OF TOBACCO CONSUMPTION Implication for tobacco production 5.18. The nicotine and tar content of a cigarette depends on a number of factors (type of tobacco used, position on stalk, burn-up rate, filling properties, filter, etc.). As each cigarette consists of a blend of several varieties, there is no direct link between the cigarette's nicotine and tar content and that of any specific tobacco used in the blend.

The consumer finds the nicotine and tar content of a cigarette important. The producer therefore has to adjust his product to consumer demand for lighter cigarettes and thus produce tobacco containing less nicotine and tar. Exports to developing countries 5.24 and 5.27. The argument that Community export refunds are responsible for a high death rate and for high health costs in developing countries is exaggerated. Export refunds are payable for all destinations, with the exception of countries producing the same type of tobacco (the United States and Canada in the case of flue-cured tobacco and Burley;

Turkey and the former Yugoslavia in the case of oriental varieties). A considerable proportion of Community exports is of high quality and sought after on the world market. Community refunds neither cause nor promote imports of low-cost tobacco by countries outside the Community. These countries want cheap tobacco and would obtain what they want on the world market even if they could not get it from the Community. 6. TOBACCO IN THE USA 6. When comparing the Community system with the American system, a number of basic differences should be noted: - The first important point is that tobacco in the United States is a major crop and as such very much protected against foreign products. - Secondly, the United States produces the higher-quality tobacco essential for the Virginia or 'American blend` type of cigarette which accounts for most of the American and international market.

The American industry is the main purchaser of American tobacco and dominates the internal market in both baled tobacco and cigarettes. The prices paid to American producers are far higher than prices in Europe (including premiums) as the American system revolves around the manufacturers' purchase commitments (accompanied by penalties in the event of non-compliance), which are not compatible with GATT commitments (Articles II and III). It is hard to imagine the Community, whose own import arrangements are consolidated under GATT, imposing such arrangements on a sector largely dominated by (American) multinationals.

The Court could have taken other producers such as China or Zimbabwe, where no support is granted to producers, as better examples for backing up its theory that the tobacco sector is too costly for the Community. Generally speaking, the situation in the Community cannot be compared with the situation anywhere else. 7. CONCLUSION 7.1. The Commission disagrees with the claim that 'the financial situation of the CMO.. is not properly presented in the Commission's consolidated accounts`: (a) The information is available at the Commission; many of the Court's findings are in fact contained in various Commission reports and memoranda. - The Court is right when it states that 'annual expenditure is not definitive until clearance procedures... have been completed`.

The Commission cannot close the accounts without first checking that expenditure has been properly incurred. - Expenditure is not 'understated`: - If there is a profit on sales in a given financial year, there will have been a loss in the preceding year or years as a result of depreciation. - Expenditure disallowed during the clearance of accounts is not deducted from the expenditure relating to the market organization but is allocated to a specific budget heading covering the adjustments for all sectors of the CAP (Chapter B1-37). - In fact, this procedure may be said to have the effect of overstating, rather than understating expenditure in each sector since the amounts recovered by the EAGGF are not taken into consideration when determining the actual cost of each market organization.

(b) The fact that advances are paid or that definitive payments are made for tobacco harvested eleven years before merely reflects the period elapsing between the chargeable event and payment. (c) Supplementary and amending budget No 3/92 allocated ECU 17 million to Article B1-173 (Other intervention for tobacco) for the precise purpose of identifying and highlighting the balances of premium payments by Greece in respect of marketing years in the distant past. These where quite distinct from the expenditure on premium payments in a normal year. As stated above, intervention expenditure on tobacco is not entered in a heading earmarked for premiums since it can be divided into two:

(a) expenditure on a premium designed to guarantee the planter a minimum price and (b) expenditure corresponding to the value of tobacco on the Community market. 7.2. The claim that there is 'no assurance that in the three main producing countries expenditure was incurred in a regular and lawful manner` cannot be based on paragraphs 3.25 to 3.69 and 3.89 to 3.117 alone (the paragraphs invoked by the Court). The Court does not specify the amount involved in each of the various infringements established. It cannot therefore be ascertained whether they have led to financial corrections under the clearance procedure or to recovery under the procedures relating to irregularities.

7.3. The Commission has done its best when reselling tobacco from intervention. - The Commission recognizes that irregularities may have been committed in the past (second sentence). The investigations which the Commission has asked two Member States to conduct will show the real extent of these irregularities and enable the lost funds to be recovered. In the meantime the Commission has taken every measure necessary to guard against any irregularities. - The Commission's knowledge of the market has increased considerably in the recent past, in particular following its sale of intervention stocks by type of tobacco and its analyses of market trends based on all the information available.

The Commission would nevertheless point out that the number of traders involved in this sector is limited; the risk of cartel information will therefore continue to exist, whatever the volume of the lots offered for sale. It will, however, endeavour to forestall any such 'market organization`. 7.4. The Commission is closely monitoring implementation of the new market organization in all the Member States. However, the EAGGF teams responsible for the inspections need additional staff. 7.5. The Commission has adopted rules to ensure that the irregularities detected are closely monitored. However, the Commission regrets that the Member States are faint-hearted in this respect and intends to ask the Council to help overcome their current reluctance to act.

7.6 and 7.7. In 1990 the EAGGF launched a general enquiry into tobacco exports from Greece and Italy. The first financial corrections were made during clearance of the 1990 accounts. The ongoing enquiry will decide whether measures should be taken in respect of earlier years. 7.8. The Commission will examine the question raised here. 7.9. Experience shows that no reliable or meaningful data could be obtained because of the large number of categories of variety. Details such as the world market price (Variety? Origin? Quality category? Analytical composition? Presentation?), processing costs, losses, etc., will no longer be needed under the new market organization.

7.10. The Commission was aware of the limits on its activities under the old market organization. This is why it proposed to a Council a new market organization which was fundamentally different from its predecessor. As regards reduction of the premiums when the maximum guaranteed quantities are exceeded, the Commission faces procedural constraints since it has to propose prices before details of the previous harvest are known. 7.11. Under the new arrangements the premium is paid once the quantities delivered to the first-stage processors have been checked and before the processed tobacco has left the stores.

7.12. (a) The purpose of all market organizations is to achieve the objectives laid down in Article 39 of the Treaty. The Court's view of the new arrangements appear to be too imbued with experience of the old market organization. Following the reform, the information obtained from the Member States and the private sector will play a less decisive role in the establishment of premiums. 7.12. (b) It is simpler and easier to check that the processor has paid the producer the entire premium than to check minimum prices based on the quality supplied since the premium is always the same and does not depend on quality.

7.12. (d) The final payment of the premium to producers is in no way dependent on the control of sales to the final consumer. 7.12. (e) The first preventive controls carried out show that the Member States have computerized all the relevant data contained in the cultivation contracts and certificates. Cross-checks are carried out to detect anomalies resulting when data on such aspects as identity, tax code, area under contract, variety, quota and production area do not tally. The computer lists have been sent to the regional offices, which check these details in advance against the producers' cultivation certificates.

The documentation referred to above will be checked once more when the tobacco is delivered. 7.13. The Commission considers that the Court's proposal to follow the American example in some or all respects does not take account of the basic differences between the tobacco sectors in Europe and in the United States. 7.14. When setting quotas the Commission will obviously take account of existing stocks, projected exports and internal demand. It would be unrealistic to impose a scheme which involved any commitment to purchase tobacco. 7.15. Community quotas are transferable in a number of cases, e.g. the sale, hire, renting or inheritance of the farm.

The fact that the transfer of quotas is restricted in other cases is due to Parliament's demand that traditional producers should continue to grow tobacco. The Commission will examine whether these restrictions can be relaxed once it has sufficient information about how the scheme is actually applied. 7.16. Now that the sector has been reformed and, in particular, now that intervention and export refunds have been abolished, Community production should be more in line with market demand. However, the support granted to European tobacco-growers in no way influences the public health costs which may result from the consumption or abuse of finished products.

7.17. Income support in the form of the per-hectare aid granted for major crops is impractical. Labour costs per hectare are low in the case of major crops but are the decisive factor in the tobacco sector. The introduction of per-hectare aid for tobacco would lead to negligence in cultivation and to a drastic fall in quality. ANNEX Brussels, 13 September 1990 Sir, Articles appearing in the press in July seriously impugned the honour of a Commission official. As it would like to be in full possession of the facts before it takes any measures which might prove necessary, the Commission wants the situation fully clarified.

I would therefore ask you to examine whether there were any irregularities or shortcomings in the way in which the Head of the Tobacco Division carried out his responsibilities. Guy Legras Mr Hans-Helmut WÄCHTER