32000Y0323(02)

Special Report No 1/2000

on classical swine fever, together with the Commission's replies

(pursuant to Article 248(4) second subparagraph, EC)

(2000/C 85/01)

CONTENTS

>TABLE>

Summary

Commission management

A total of ECU 570 million for exceptional market support measures and ECU 80,5 million for veterinary measures has been charged to the 1997 and 1998 budgets (see Table 1).

The Commission's action in the veterinary sphere is essentially to ensure proper application of Community legislation, and of Directive 80/217/EEC in particular. However, the recent epidemic has highlighted the need to modify the Directive in order to take account of experience and changes in scientific knowledge.

In the veterinary sphere, eradication and surveillance plans have not kept the Community free of classical swine fever (CSF) and the Commission did not ensure that Member States submitted contingency plans for dealing with CSF after the 1993 to 1995 epidemic. The Community's animal movement notification system (Animo) has failed to provide vital information at crucial times and must be modified in order to enable Member States to use it when fighting epidemics. The Community's non-vaccination policy, which is based on the nature of vaccines and the cost of a vaccination programme compared to eradication, is called into question by the development of marker vaccines (not available on the market as at the end of 1998), and the very high costs associated with the 1997 to 1998 epidemic. The Commission should update its cost-benefit analysis and use it for the debate on policy changes.

The unit responsible for the management of agricultural expenditure carried out several inspections in the Member States, as a result of which payments to the Netherlands have been reduced by ECU 20 million for alleged breach of Community legislation (see paragraph 39) and the balance payable to Spain may be reduced by up to ECU 6,6 million for ineligible expenditure (see paragraph 41). The clearance of accounts unit has also carried out controls in the Member States, but has yet to determine any financial consequences on conformity grounds.

Under exceptional market support measures the Management Committee approved successive increases in the initial ceilings from 1,1 million to 10,8 million animals (see paragraph 31). A number of categories of pigs and weight bands were introduced, as well as specific schemes (canning and insemination ban), at the request of the Member States concerned.

Implementation by the Member States

Particular problems have been encountered with regard to preventive killing. In the Netherlands, the temporary suspension of preventive killing is believed to have contributed towards the spread of the disease and the subsequent cost to the Community budget of at least ECU 135 million (see paragraph 49). In Spain the use of "partial" preventive killing cannot be regarded as a veterinary measure (see paragraph 51).

The Dutch "overname" (buying-up) scheme was poorly administered (see paragraph 56), the healthy status of the pigs concerned and the voluntary and partial nature of the scheme mean that ECU 70 million included in the claim for Community financing under veterinary measures has been rejected by the Commission as ineligible (see paragraphs 63 to 65).

Payments to producers cannot be verified in a large number of cases because valuations were not adequately documented (see paragraph 53), the weights of the animals were often estimated and animals buried on farm were not weighed (see paragraph 59). Rates of compensation for certain categories of animal such as gilts and sows were not always documented (see paragraph 61).

The use of municipal pits to bury 140000 pigs in Spain (see paragraph 50) and the use of cold stores in the Netherlands meant that it was not always possible to trace batches of pigs from the farm to the rendering plants/cold stores. The problems with regard to tracing batches of animals through to final destruction are common to the market measures, but are more serious given the number of animals involved (see paragraph 57).

Conclusion

Since the 1993 to 1995 epidemic, progress has been slow in developing a specific strategy to tackle subsequent outbreaks of the disease. In the meantime the Community has had to face its worst ever epidemic of CSF. Emphasis on other problems in the veterinary sphere (establishment of the free internal market and BSE) meant that CSF contingency plans were not given the priority the 1997 to 1998 epidemic demonstrated that they deserved. The Court urges the Commission to dedicate the resources necessary to complete the revision of Community legislation and policy in this area as quickly as possible. Member States must also assume their responsibility for effective implementation of Community legislation in order to reduce the risk of future outbreaks of CSF causing epidemics on such a scale.

Introduction

Principal features of the market

1. The common market organisation for pigmeat traditionally provides only limited market support for the sector, but when there is an epidemic of classical swine fever (CSF) Community expenditure on exceptional measures to support the market and veterinary measures results in a quantum increase in the charge to the budget (see paragraph 37). A total of ECU 570 million for exceptional market support measures and ECU 80,5 million for veterinary measures has been charged to the 1997 and 1998 budgets (see Table 1).

2. EU production, is around 16,6 million tonnes per annum. The pig population has remained relatively stable during the 1990s at around 119 million head. The removal of 12 million pigs (see Table 2) from the market during the 1997 to 1998 CSF epidemic led to a reduction in supply equal to 1 million tonnes of pigmeat. Consequently, there were substantial price increases in the first six months of 1997 and the average Community market price for pig carcases remained very high (ECU 164/100 kg) during the whole of 1997.

3. Per capita consumption has been increasing steadily and is predicted to continue along this trend (see Table 3). The EU is self-sufficient in pigmeat production (1994 = 106 %(1)) and is the world's largest exporter. In order to ensure access to world markets, pigmeat must come from an area with a "high health status" as defined by the International Office of Epizootics, i.e. free from, inter alia, CSF.

4. Prices saw a dramatic upsurge during 1996 (13 %) partly as a result of the BSE crisis (consumers switched from beef to pork). During 1997, prices also increased as a result of reduced supply due to CSF. In 1998, prices have fallen back just as sharply due to a combination of factors - greatly increased domestic supply and a decline in third country exports - which have triggered private storage aid(2).

5. Community production is highly concentrated - 70 % of the pig population is in five Member States, 21 % in Germany alone (see Table 4). Within these Member States regional concentration results in huge numbers of pigs being kept in close proximity. In Spain, for example, one third of the total pig population is in Catalonia, 95 % is in Vlaams Gewest in Belgium, 90 % in the east and south of the Netherlands and 65 % in the three Länder of Bavaria, Lower Saxony and North Rhine-Westphalia in Germany (see Table 5). For all these areas, with the exception of Catalonia, livestock density exceeds the level that is generally accepted as ecologically sustainable (1,4 livestock units per hectare)(3).

6. More generally pig production has become highly specialised, piglets are bought in as cheaply as possible and fattened on a standard diet to a specified weight within strictly determined production cycles and are then sold on the open market for EU consumption or for export. This is only profitable on an increasingly larger scale as margins are squeezed as a result of competition at home and abroad. This has resulted in significant increases in the average number of pigs per holding and the proportion of large holdings. During the period 1990 to 1995 the average number of pigs per holding in Belgium increased by 68 %, in Spain by 59 % and in the Netherlands by 36 %. The increase in Germany was less marked (19 %) due to the restructuring that took place after unification (see Table 5).

7. Specialisation in Member States such as the Netherlands results in the movement of piglets from farm to farm, whereas closed systems, such as the one dominant in Denmark, minimise the number of movements and thereby the risk of spreading disease. The Dutch authorities are well aware of the consequences of high-density pig farming and have recently passed legislation designed to reduce the number of pigs in such areas. A system of pig entitlements has been introduced from 1998 and imposes a general 10 % cut on the number of pigs kept per holding. As of 1 January 2000 the pig entitlements allocated for 1998 would have to be reduced by 15 %. In addition the financial impact on the State of any future outbreaks of CSF would be mitigated by a levy on pig producers introduced in 1998 to cover the costs of combating animal diseases on pig holdings. However the legislation has been successfully challenged in the courts by pigmeat producer organisations. Early in 2000, the Dutch Government will present new draft legislation to the Dutch Parliament for approval. According to the Dutch authorities, this new proposal will take account of the outcome of the court proceedings.

The scope of the Court's audit

8. In the 1995 Annual Report the Court reported on measures taken by the German and Belgian authorities to combat the 1993 to 1995 epidemic of CSF. For the 1997 to 1998 epidemic the Court's primary objective has been to evaluate the management of the crisis in the four Member States affected and at Commission level.

Table 1

Expenditure 1997 to 1998

>TABLE>

Table 2

Number of pigs killed under exceptional market support and veterinary measures

>TABLE>

Table 3

EU pigmeat supply balance

>TABLE>

>TABLE>

Table 4

Total of the EU pig population (1993 to 1997), December

>TABLE>

Table 5

Regional concentration of pigmeat production in Belgium, Germany, Spain and the Netherlands

>TABLE>

Table 6

On-the-spot controls concerning CSF measure (1997 to 1998)

>TABLE>

9. As well as visiting the Member States the Court has visited the Community CSF Reference Laboratory in Hanover. Many of the Court's observations in this Report reflect the findings of the Commission's own control services, with which the Court has worked in close cooperation. A list of all the visits made in this context by the Commission and the Court during 1997 to1998 is in Table 6.

Commission management

Combating classical swine fever

10. The pig is the only natural host for the classical swine fever virus. The disease has never been reported in human beings. Depending on the virus strain it can cause a high degree of mortality among infected pigs. The main symptoms are a rise in body temperature, depression, loss of appetite, conjunctivitis, stilted gait, incoordination and diarrhoea. CSF has an average incubation period of seven days, is highly contagious and easily transmitted. By the end of the 1980s the EU wild boar population was CSF-free but has since become infected again, possibly as a result of eating contaminated leftovers. CSF in wild boar does not die out naturally and is now endemic in the wild boar populations of Germany, Italy and a part of France. Wild boar transmit the virus to domestic pigs.

11. CSF may be tackled on two fronts: prevention and dealing with outbreaks after they have occurred. On the preventive front the Community has three basic instruments: eradication and surveillance programmes primarily aimed at controlling the spread of the disease in wild boar, contingency plans and the possible use of vaccination. Once an outbreak has occurred protection and surveillance zones are put in place, the pigs on the infected holding are killed and movement restrictions remain in place for fixed periods. In the case of multiple outbreaks movement restrictions may be in place for relatively long periods, thereby preventing producers from marketing their pigs. In these circumstances the pigs are purchased under exceptional market support measures. Monitoring the movement of pigs across borders from farms that may turn out to be suspect areas is essential to avoid the epidemic spreading in this way. The animal movement system (Animo) is used for this purpose. The veterinary and market support measures are subject to on-the-spot controls by various Commission services.

12. Since 1990, there have been 1445 outbreaks of the disease in the EU (see Table 7). These outbreaks have increasingly occurred in areas of high density. The traditional instruments of control developed to tackle the problem in low-density areas are less effective in high-density areas. High-density concentration and specialisation combine to pose a threat to the environment and increased risk of contagion in the event of epidemics such as CSF. The Commission believes that the potential for effective disease control in such situations is likely to deteriorate unless livestock density is reduced(4).

Table 7

Number of outbreaks of CSF

>TABLE>

Eradication and surveillance programmes

13. The most recently approved programmes are for the eradication of CSF in feral pigs in Varese and Brandenburg, Mecklenburg-Western Pomerania and Lower Saxony(5). Austria, Germany and Italy are the only Member States that have had eradication and surveillance programmes for CSF in domestic pigs during recent years. The Netherlands did not have such a programme, the last recorded cases before the 1997 epidemic were in 1992.

Contingency plans

14. The Commission has stated that for outbreaks of diseases of major importance such as CSF "the responsibility for preparedness rests primarily with the central veterinary administration of the Member States, who through well rehearsed contingency plans should be prepared to meet any challenge"(6).

15. Council Directive 80/217/EEC(7) (Article 14) required Member States to submit CSF contingency plans by 1 January 1993 for Commission approval. The Commission "shall examine the plans in order to determine whether they permit the desired objective to be attained... [and] ensure they are compatible with those of other Member States". The plans from Belgium, Spain and the Netherlands, submitted to the Commission only in December 1997, were not approved until March 1999(8). The Commission should have ensured that Member States presented programmes immediately after the 1993 to 1995 epidemic and that tests were carried out, especially for those with a recent history of the disease. Failure to do so must be regarded as one of the reasons why the Animo shortcomings (see paragraph 24) were not identified sooner and for the diversity of measures taken by the Member States.

Vaccination

16. In 1977, the Commission published a paper by P. R. Ellis et al. entitled "Studies on the epidemiology and economics of swine fever eradication in the EEC"(9) that prompted the decision taken in 1980 to eradicate CSF by adopting a common policy of slaughter, together with the gradual abolition of routine vaccination. Council Directive 80/217/EEC(10) laid down the procedure to be followed in the event of an outbreak of CSF. By 1990, all vaccination had ceased.

17. The main arguments against the use of vaccination at that time were based on the nature of the vaccines themselves (live-virus type vaccines require careful handling) and the cost of vaccination, given the rapid turnover (reproductive cycle) of the pig population.

18. The problem with traditional live vaccines is that no serological distinction is possible between animals after vaccination and those convalescent from natural infection. This has meant that if a third country imported pigmeat from a Member State which had used vaccines, when the meat was tested it could either show CSF infected or simply be meat from a vaccinated animal. The development of marker vaccines may overcome this fundamental problem. Whilst the Scientific and Veterinary Committee fully supports the EU non-vaccination policy, it endorses the restrictive use of marker vaccines in densely populated pig areas as a means of reducing the virus spread and concomitant economic losses. Marker vaccination is to be one of the tools used to eradicate the disease and should not be used as a preventive measure. As such it would provide an alternative for pre-emptive slaughter and destruction of animals in "apparently uninfected neighbourhood farms (mass slaughtering is expensive, ethically questionable, wasteful of food resources and may destroy genetically valuable animals)"(11). A recent trial involving the use of two marker vaccines, for which an application for marketing authorisation had been made to the European Agency for Evaluation of Medicinal Products (EMEA) is currently under evaluation.

19. No cost-benefit analysis of the use of vaccination has been performed at Community level since1977. Since then the cost of eradication has increased substantially, especially in high density areas. The Court calculates that the total direct cost (i.e. Community and national) of the epidemic in the Netherlands (excluding veterinary costs for 1998 not claimed as at that time of the audit) amounts to ECU 1075 million. In Belgium the cost per pig killed was approximately ECU 115, which is broadly in line with the cost per pig killed during the previous epidemic (ECU 100) whereas for the Netherlands it amounted to ECU 157 per pig killed. The Commission should undertake a new comprehensive cost-benefit analysis of alternative control and eradication strategies.

Dealing with outbreaks

20. Under the terms of Council Directive 80/217/EEC of 22 January 1980 (as last amended by Council Decision 93/384/EEC)(12) which introduced Community-wide measures to control CSF, Member States must notify all suspicion of outbreaks of CSF. The Standing Veterinary Committee decides what action to take in the event of a potential epidemic. This generally takes the form of a Commission Decision banning the movement of pigs from specified areas of the Member State affected. These decisions are modified and repealed as the situation evolves.

21. Suspicion of CSF must be notified to the competent authority and the holding placed under official surveillance - no pigs may enter or leave the farm. Surveillance measures may not be lifted until CSF has been officially ruled out. If CSF is confirmed, all pigs on the holding must be slaughtered without delay and be destroyed. Any material that is likely to have been contaminated must be treated in order to ensure the destruction of the virus. The farm must be cleaned and disinfected and may be restocked 15 days afterwards. A detailed epizootiological report is required for each outbreak. As soon as CSF has been confirmed on a holding, a protection zone with a radius of at least three kilometres around the outbreak site should be established. This protection zone forms part of the surveillance zone which is established at the same time and must have a radius of at least 10 kilometres around the outbreak site.

22. Within the protection zone pigs may not be moved from holdings until 21 days after the disinfecting of the outbreak holding. The measures applied in the surveillance zone are similar in nature but the restriction applies for seven days after disinfecting. The protection zone may be lifted 30 days after completion of the cleaning of the infected holding.

23. The implementation of these measures is the responsibility of the Member States but is, to a large extent, dependent from the outset upon the ability of Member States to trace consignments of piglets/pigs that might originate in zones affected by CSF in order to test them and/or eliminate them in order to prevent an epidemic.

Notification of animal movements

24. Council Directive 90/425/EEC of 26 June 1990(13) concerns veterinary and zootechnical checks applicable in intra-Community trade in certain live animals (including pigs in general) and products with a view to the completion of the internal market. This Directive is the legal basis for the computer-based animal movement system (Animo), the ultimate aim of which is to ensure that veterinary checks are carried out at the place of dispatch only. Until this aim is achieved, Member States are permitted to carry out spot checks at the place of destination. Whilst the Community provided financial assistance to Member States in order to enable them to acquire the necessary computer equipment, the Commission does not manage the system - this is done by a private company.

25. Article 3 requires, inter alia, the Member State of origin to ensure that animals that might have to be slaughtered under a national eradication programme or cannot be marketed on their own territory for health or animal health reasons are not dispatched to another Member State. The purpose of the destination spot checks is to ensure that all the requirements of Article 3 have been met.

26. According to an analysis of the causes of infection during the 1993 to 1995 epidemic in Germany(14), whereas more than half of all primary outbreaks were due to contact with wild boar, more than a quarter of secondary outbreaks are due to trade in pigs. The Spanish authorities have stated that one of the main causes of the spread of the disease in Spain was the failure of the Animo system.

27. In Belgium the veterinary service has carried out a comparison between the information recorded on Animo and that from other sources. This showed that the Animo system suffers from three significant weaknesses:

(a) incompleteness of the recorded imports; only 81 % were notified via Animo;

(b) delays; in order for swift action to be taken in the case of, for example, CSF-suspect pigs, notification must take place at the latest on the day of delivery; only 7 % of Animo notifications were received within this time;

(c) inaccuracy; the number of pigs stated on the Animo notification compared with the actual number delivered (confirmed on farm or using other sources) differed in 26 % of cases.

28. Whilst there is no such comparative data for the crucial period during which imports from the Netherlands introduced CSF to Belgium, there is no reason to believe that the system was performing any better at that time. A concrete example of the failure of the system was detected in one of the veterinary payments examined by the Court. A consignment of 76 pigs imported from the Netherlands for which there was no Animo notification was delivered on 6 February 1997.

29. The Commission states that Animo was not intended to be used by Member States to carry out 100 % checks at destination. Nevertheless it should contain the whole population of movements from which national veterinary services may select consignments for on-the-spot checks. Clearly where health or animal health is at stake, such services should be in a position to identify all risk consignments as soon as, for example, CSF is declared in a Member State from which pigs are moved onto their territory. Animo is the only Community-wide system that can fulfil this function and the Commission should take the opportunity to enhance the speed, accuracy and completeness of the system.

Exceptional market support measures

30. Shortly after the veterinary measures were introduced in each Member State, market support regulations were adopted: for the Netherlands and Germany on 3 March 1997, on 1 April 1997 for Belgium and on 22 May 1997 for Spain(15). These regulations were repealed in accordance with progress achieved on the animal health side. Market support measures were closed for Germany in October 1997 but a special measure for piglets was later introduced(16). In Belgium the measures ceased in November 1997. The Netherlands was not clear until July 1998 and Spain until February 1999.

31. All regulations were discussed in the Management Committee for pigmeat. Initial ceilings were set for the number of animals for each category that could be bought up under the measures; these were increased at the request of the Member States from 1,1 million animals to 10,8 million (see Table 2). The market support regulations were amended a total of 44 times, primarily to reflect the spread of the disease (increased ceilings) but also to change the categories/weights of animals eligible for support and the rates of aid payable. In Spain special rates of aid were introduced for cull sows and Iberian pigs. In the Netherlands a category of very young piglets was added, an insemination ban was introduced (see paragraph 74) and cull sows were also foreseen for market support. Germany was granted approval for canning some of the meat (see paragraph 71).

On-the-spot controls by the Commission

32. The Food and Veterinary Office (FVO) of the Commission (DG XXIV) has the right to inspect implementation of Community veterinary legislation. In view of the size of the epidemic in the Netherlands it made repeated visits there.

33. After the Netherlands had declared 13 outbreaks, on 14 February 1997 a Decision(17) was taken restricting the movement of pigs from specified areas. The FVO responded quickly by visiting the Netherlands in the same week. At this juncture the inspectors concluded "It is likely that there will be some further outbreaks of CSF as the result of contacts. However, the overall situation at the time of the mission appeared to be promising with the Netherlands services taking good control measures which were well organised". They also noted problems with collection centres, tracing and the animal movement notification system (Animo) due to the inadequacy of destination addresses provided by the exporter.

34. The FVO returned to the Netherlands in April 1997, by which time there had been 65 outbreaks. It was noted that the transmission of the disease via contaminated semen had not been given due consideration by anyone and that not using preventive killing in the case of contacts was giving the virus further opportunity to spread. In June 1997, the inspectors observed that elimination of entire herds could only be applied within certain limits due to shortages in rendering capacity, the insemination ban had been introduced too late to have an impact on the total number of animals in the area when it mattered and that the market regulations were encouraging producers to keep pigs and to force their growth in order to achieve a good price. These aspects are considered under implementation by the Member States.

35. The Commission's action in the veterinary sphere is essentially to ensure proper application of Community legislation and of Directive 80/217/EEC in particular. However, the recent epidemic has highlighted the need to modify the Directive in order to take into account experiences and changes in scientific knowedge. Moreover, measures taken by national authorities, such as the Dutch decisions to suspend preventive slaughter (see paragraph 49) and to buy up pigs for welfare reasons ("overname" see paragraph 63), are not referred to the Commission for approval.

36. The unit responsible for the management of agricultural expenditure also carries out inspections in the Member States. They cover some of the same ground as the FVO, indeed some are performed jointly, but have a considerable financial element as well. Finally, exceptional market support expenditure is subject to clearance of accounts controls.

Community financing

37. Expenditure charged to the Community budget in 1997 and 1998 in respect of traditional expenditure on the pigmeat common market organisation was ECU 142 million whereas expenditure on exceptional market support measures (70 % co-financed) alone was four times greater (ECU 570 million). Expenditure met from the veterinary fund in 1997 and 1998 (50 % co-financed) in respect of 1997 measures was ECU 80,5 million. A summary of expenditure for 1997 and 1998 is in Table 1. Payments from the veterinary fund may take the form of reimbursements or advances. Expenditure on CSF is included under the general heading for emergency veterinary measures and as such should be separately identified in the budget given the substantial sums involved.

38. Belgium and the Netherlands received a first payment in respect of 1997 expenditure in December 1997. Other payments were made during 1998 but the balance in respect of 1997 will not be paid until 1999. This is partly due to a shortage of appropriations, but also reflects the fact that a considerable amount of expenditure may not be accepted by the Commission. If the claims are met in full, the amount payable by the Community is ECU 49 million in respect of 1997 and ECU 34 million in respect of claims for 1998.

39. The Netherlands presented a claim for NLG 195 million (ECU 88 million) on 26 September 1997 covering the first 217 outbreaks of CSF in 1997. The Commission deducted 6 % for VAT (ineligible) and applied a flat-rate deduction of 25 %, (some ECU 20 million) and paid ECU 31 million in respect of the first 195 outbreaks only. The 25 % penalty was based on the failure to apply Article 3 of Council Decision 90/424/EEC of 26 June 1990 on expenditure in the veterinary field(18):

(a) the imposition of suitable measures to prevent the risk of the spread of infection;

(b) the slaughter of animals of susceptible species;

(c) adequate compensation of producers;

(d) the creation of protection zones.

40. On 22 June 1998, the Dutch authorities submitted a "new" claim for the whole of 1997. The amount of the claim was NLG 893605 068 (ECU 402 million) for 50 % co-financing. The claim covers 424 outbreaks and also includes NLG 153 million for the "overname" measure. A second payment (ECU 32,4 million) was made in January 1999 in respect of outbreaks Nos 196 to 360(19). Due to shortage of appropriations a further ECU 11 million payable under the same Decision was paid from the 1999 budget.

41. On 2 October 1997, Spain submitted a provisional list of payments in support of its claim for financial aid for veterinary measures under the proposal that later became Commission Decision 98/63/EC of 9 December 1997(20). This covered the first 57 outbreaks in Catalonia, Castilla y León, Murcia and Andalucía. According to the Decision, in order to receive the advance, Spain had to submit an epidemiological report for each and every outbreak and contact holding and a financial report. On 16 June 1998, a "final" list was submitted together with other documents required under the terms of the Decision. Rather than wait until all the supporting documentation had been submitted, the Commission considered it had sufficient grounds to pay a ECU 4 million advance in January 1998 which was charged to the 1997 budget. The Commission may refuse up to ECU 6,6 million in connection with certain measures it regards as ineligible (see paragraph 51).

42. The Guarantee Fund expenditure includes a charge of ECU 6,8 million to the 1997 budget as a result of Commission Regulation (EC) No 76/97 of 17 January 1997(21). Initially the slaughter of 90106 pigs in the Damme region of Germany between December 1993 and March 1994 was presented as a veterinary measure but was rejected for Community financing because not all of the criteria for such measures had been met. Under this Regulation the purchase of 63074 of these pigs (70 %) is treated as "exceptional market support measures" to be borne by the Community budget, i.e. ECU 2 million more than it would have cost had it been accepted as a veterinary measure. A similar practice had been adopted in respect of Belgium in January 1996(22) for compensation paid for the purchase of 40633 pigs in December 1993: the Court criticised this procedure in its 1995 Annual Report (paragraph 3.26). It is not clear why it took a further year to reach the same conclusion for Germany. The reply given by the Commission referred to the grey area between veterinary and market support slaughtering. The problem of the grey area remains and is the cause of much debate between the Commission and the Member States.

Implementation by Member States

Veterinary measures

Origin of the 1997 to 1998 epidemic

43. The epidemic started in 1996 in Paderborn in Germany. The producer concerned had a swill licence, but in order to save money he did not heat the swill sufficiently to kill off the virus. Charges have been brought against the two parties suspected of being responsible for the outbreak of CSF. In addition, the Paderborn Public Prosecutor's Office opened preliminary criminal proceedings against a veterinary surgeon who was accused of spreading the swine fever virus due to inadequate precautions in the course of treatment of pigs on various farms. It dismissed the possibility of criminal proceedings but his licence to practice has been revoked. From Germany the virus spread to the Netherlands (no proof of method) and thence to Belgium and Spain via the movement of infected animals(23).

44. In Spain no cases of CSF had been recorded for almost 10 years until February 1997. During the period 1 to 24 February 1997, some 220000 piglets and 19000 adult pigs were imported from the Netherlands. Of the 354 consignments concerned, 212 originated in affected regions of the Netherlands (134000 pigs). The Animo system did not work during the period 1 January 1997 to 17 February 1997, it was therefore not possible for the Spanish authorities to trace a large number of the animals imported (see paragraph 26). Two farms were however identified as having received such animals and the herds were killed as a precautionary measure, although there is no documentary evidence to support the action taken on 24 February on one of the farms. Although the first suspicion was recorded on 14 April 1997, the Commission concludes that the CSF virus was active in Spain from February to April 1997 without being detected. The first two outbreaks occurred on holdings belonging to the same group as one of the two farms cleared in February. This group, according to the Commission, must be held responsible to a large extent for the spread of CSF in Spain. The group experienced five outbreaks and 13 preventive slaughter operations and received a total of ECU 9,2 million from Community funds.

45. The first Belgian outbreak was confirmed on 30 June 1997. All but one of the later outbreaks are directly connected with it. Because the farmer is suspected of late notification of the disease, the authorities prudently decided to impose a 50 % cut on the compensation paid and disciplinary procedures have been started against the farmer's veterinarian. The farmer has taken legal action.

Preventive killing

46. Member States are not required to use preventive killing but if they do so, the policy should be soundly based from both the technical and scientific points of view. Belgium and Germany made use of preventive killing during an earlier epidemic and it was found to be one of the more effective ways of tackling the problem in high-density areas. With regard to the 1997 to 1998 epidemic, Spain and the Netherlands chose to make use of preventive killing for this reason; in Germany it was not considered as practicable, whereas Belgium applied it in the context of Article 6 of Council Decision 90/424/EEC to deal with the threat posed by CSF on neighbouring farms in the Netherlands.

47. From the beginning of the CSF epidemic the Dutch authorities adopted a policy of preventive clearing of farms within one kilometre of the first two confirmed outbreaks in order to reduce the risk of the disease spreading to surrounding farms. In February 26 farms were cleared (26000 pigs) three of which were later found to have been outbreaks (two classified as neighbourhood contacts). Scientific opinion was divided at the time, some supported the policy of preventive killing, others were against it. On 14 February, on the basis of advice given by national experts, the Dutch authorities decided to suspend preventive killing on the grounds that it was obstructing the collection of important epidemiological information; preventive killing was resumed two months later. During this period market support measures were in operation, under which 690000 animals were bought up.

48. By March 1997, it was known that 39 farms must already have been infected in early February (conclusion of epidemiological investigation). Had this been known at the time the measure should have been applied quickly to all these farms. By the end of the suspension period there had been 91 outbreaks and it was obvious that preventive clearing was essential, but the shortage of rendering capacity meant that preventive clearing could only at first be carried out within a sub-optimal radius of some 500 metres and was delayed by up to 14 days following the decision to clear, thereby undermining the effectiveness of the strategy. Of the 1182 farms preventively cleared 67 were later classified as outbreaks. It was not extended to the desired one kilometre radius until 12 May 1997. A million pigs were killed under the preventive killing measure.

49. According to a simulation carried out by the Landbouwuniversiteit in Wageningen(24), had a policy of preventive killing been rigorously maintained, far fewer farms would have had to be preventively cleared and the number of outbreaks would probably have been around 70. On this basis the total associated compensation costs would have been ECU 135 million lower and subsequent market support costs would have been lower. The Commission and the Court share the view that once a policy of preventive killing has been adopted, it should be consistently applied and that the Dutch decision to suspend it has had an impact on the development of the disease and has consequently increased expenditure.

50. In Spain a decision was taken on 7 May 1997 at a local "CSF experts" meeting whereby all farms within a radius of 250 metres of an outbreak were to be cleared of all pigs except reproductive animals (sows and boars). Weekly veterinary checks on the animals' temperature were required, but the Commission's inspection revealed that this was not systematically carried out. Some 20 % (ECU 6,6 million) of the total veterinary compensation claimed by Spain relates to "partial" preventive killing, approximately half of this was paid to producers who later received compensation under market support measures. Due to a lack of rendering capacity the authorities were unable to meet the demand from producers to sell animals under the market support measure, so they decided to open 11 municipal pits in which to bury the carcases (in all some 140000 pigs were buried in these pits). This meant that what would have been a market measure could not be treated as such and was financed as a veterinary measure and labelled "preventive". Producers were paid more under the latter system.

51. The justification for partial preventive killing on veterinary grounds is, at best, debatable. Moreover operations on successive days were treated differently - veterinary one day, market the next. The Spanish authorities claim that the number of outbreaks declined rapidly as a result of the introduction of the partial slaughter policy, but the number of outbreaks was already in steep decline before the measure was introduced. Partial preventive killing has proved more expensive, in many cases it would have been far cheaper to slaughter the entire herd as a preventive measure rather than to partially clear the farm and later to buy up animals under market measures - saving ECU 300 per sow. The Court shares the Commission's serious doubts about the eligibility of expenditure claimed under this measure.

52. In Belgium, preventive killing is considered to be an effective way of reducing the risk of the disease spreading and has become standard policy. All pigs on farms within a radius of one kilometre of an outbreak are killed: six farms (3647 pigs) close to the Dutch border were preventively cleared. This type of measure is foreseen under Article 6 of Council Decision 90/424/EEC. However, Belgium has also made use of partial preventive killing under Article 6 of Council Directive 80/217/EEC which does not explicitly refer to partial preventive killing. Batches of risk animals - those imported from the Netherlands (16000) and sows (100) inseminated with suspect semen imported from the Netherlands - were killed. Based on the assumption that the virus had not spread in the Netherlands before 15 January 1997, it was decided that imports after this date should be treated as risk animals. However, only the batches of pigs originating in suspect areas were killed other batches that were transported on the same lorry but originating from a CSF-free area in the Netherlands were not killed.

Compensation

53. In the Netherlands standard compensation rates were set by the authorities for some categories of pig but these were generally overridden by the valuer. Explanations of higher rates were not documented and ex post statements by the valuers concerned did not fully justify the amounts paid.

54. For certain categories of farms, the rate of aid payable for sows fixed by the Dutch authorities was based on a rate per sow plus an assumed number of 4,73 piglets. This led to some producers being paid compensation for many more piglets than were on farm at the time of killing. In one of the cases examined by the Court, one producer was effectively paid for 1272 piglets more than he had. The Commission has calculated that producers were unduly paid for 175000 piglets the financial consequences of which would have been an additional charge to the Community budget of ECU 10,7million had not this expenditure been rejected by the Commission. The Commission also has evidence that some of these "missing" piglets may have been sold under market measures shortly before the sows were eliminated under preventive killing. In such cases the producers concerned were effectively compensated twice for the same piglets.

55. In the Netherlands if an incorrect amount has been entered on the valuation form but the correct amount is apparent from the valuer's notes the legal viability of the claim depends on whether the farmer could have had official knowledge of the notes. The payment calculation was performed manually and compensation by type of animal was not data-captured. In some cases a comparison between the weekly guide prices and the values used for payment revealed a difference of more than 20 %.

56. With regard to the "overname" scheme (see below), the review by a private audit firm (Moret, Ernst and Young (MEY)) revealed overpayments totalling some ECU 340000 had been made. The examination of a sample of transactions by the Commission and the Court also revealed a number of other overpayments including one case where compensation was paid twice for the same piglets ("overname" and preventive).

57. For those animals sent to central killing locations there is no possibility of following the audit trail until rendering. In their analysis MEY have tried to reconstruct this trail. MEY concludes that the 33 % difference between the 47,9 million kg taken in at the central killing places and the 31,9 million kg taken in at the rendering plants can be adequately explained by the fact that many of the carcases were subjected to a process to remove moisture prior to rendering. The Dutch authorities have also undertaken to reconcile total intake and output for each killing location.

58. In Belgium valuation is performed by an expert valuer. The number of animals of each category (weaner, non-weaner, sow, gilt, fattening pig, breeding boar) is determined, and the average weight of gilts, fattening pigs, and weaners is estimated. For mixed batches the actual weights were not taken into account, although they could have been used to produce a more soundly-based payment calculation. In three cases of mixed batches audited by the Court the estimated weight used to calculate the compensation was above the actual weight by as much as 18 %. Because of the degressive kg tariff this resulted in overpayments of a lower magnitude.

59. Similarly in Spain, different categories of animal were weighed together. It is therefore not possible to confirm that the correct amount of compensation has been paid. This was invariably the case for 67500 animals buried on farm, for which veterinarians' estimates of weights were used to calculate the aid. The Court established that overpayments totalling ECU 1,2 million occurred as a result of failure to apply the correct rates of compensation.

60. In Germany a valuation committee establishes the value of the pig herd. Compensation is based on the estimated weight of the live pigs and is not adjusted in line with the actual weight of carcases received at the rendering plant. In the sample examined the live weight estimate was up to 13 % higher than the carcase weight.

61. In Belgium no documentation existed to explain the different rates used for partial slaughter and total clearing for sows and piglets. The compensation rate initially set for gilts was increased after protests by farmers. In some cases this resulted in a doubling of the compensation payable. Gilts are destined to become reproductive sows and are therefore inherently more valuable. On average, gilts received 75 % higher compensation than fattening pigs, from which it is clear that the producer has a financial interest in maximising the number of gilts rather than fattening pigs. Moreover, the point at which a piglet becomes a gilt is not clear. Animals as light as 25 kg have been classified as gilts for compensation purposes whereas under market support measures animals weighing less than 26 kg were classified as piglets. According to the document that describes the compensation procedure, on a normal production farm the number of gilts is at most 20 % of the number of sows. In specialised farms the percentage may be higher of course. During the outbreak in Belgium, on farms that were cleared, the percentage of gilts was close to 60 % i.e. three times the norm.

62. In Spain, for certain categories of animal, the difference between the aid paid under veterinary measures and under market support is significant. For example, veterinary aid for 15 kg piglets killed on 14 July 1997 was ESP 8575, whereas market support aid was 17 % higher. Both rates were more than twice the actual market price. In the Netherlands the average rate of aid for fattening pigs killed under the "overname" scheme was 20 % higher than under the market measure. The magnitude of such differences cannot be justified.

The Dutch "overname" scheme

63. The market support measures introduced under Commission Regulation (EC) No 413/97 of 3 March 1997(25)were confined to fattened pigs and piglets, but all categories of pig were subject to the transport ban. Specialisation (see paragraph 7) meant that overcrowding developed quickly and caused welfare problems on many of the farms concerned.

64. This problem was aggravated by the fact that 1675 holdings had to be declared suspect, due to the infection in two artificial insemination centres. At the beginning of the epidemic, spread of CSF via artificial insemination was not considered a real possibility. Subsequent events pointed in that direction and the approach adopted assumed that there was such a link(26).

65. In March 1997, the Dutch authorities therefore decided to introduce a nationally funded veterinary support measure (the "overname" scheme) under which producers were compensated for healthy pigs they chose to have killed. Nevertheless, in June 1998 the Dutch authorities included some ECU 70 million in their claim for Community co-financing. Based on the findings of a joint audit (September 1998) by the Court and the Commission, in May 1999, the Commission rejected this part of the Dutch claim on the grounds that the pigs concerned were healthy and thus do not qualify under the terms of Article 3 of Commission Decision 90/424/EEC and that the partial and voluntary nature of the scheme does not meet the requirements of Council Directive 80/217/EEC.

Exceptional market support measures

Purpose

66. The purpose of these measures is to remove pigs and piglets in order to avoid overcrowding on farms located in protection and surveillance zones where restrictions on the movement of animals which apply as a result of veterinary measures threaten serious disturbance of the pigmeat market. Although the animals concerned are considered to be CSF-free when they are taken off-farm, the carcases must be processed into products not intended for human consumption, i.e. rendered. The price paid to the producers for fattened pigs is equal to the market price of fattened pigs over 120 kg in weight. Lighter pigs are bought at an abated rate and piglets are bought at a fixed price per weight category. The weight is taken as the average weight of the batch. Two measures were adopted that are an exception to this procedure - canning in Germany and the insemination ban in the Netherlands.

Audit trail

67. In Belgium there is an adequate audit trail and the payment files are generally well documented. The files examined in Catalonia contain sufficient information to check that the payments have been correctly calculated. However the Commission has observed that in 10 % of the sample examined the number of animals removed according to the farm register was greater than the number subsequently claimed. Using the recorded number would have lowered the average weight of the batch, leading to either a smaller payment or rejection of the claim. The Spanish authorities did not require producers to submit copies of the registers with their claims.

68. In Germany evidence of rendering, as distinct from evidence of delivery to the rendering plant, was lacking. In the Netherlands, (Moret Ernst & Young) MEY issued an audit report dated 30 January 1998, related to the accredited paying agency LASER for 1997. In the general conclusion MEY expressed a non-opinion for expenditure related to the market measure, because LASER could not provide a global assurance that all pigs that were bought up were finally rendered or placed in cold stores. Consequently LASER introduced a new system called "VAS" to record all pig movements that took place under the measure. All relevant data from the supporting documents were registered retrospectively on the system. After an examination of that system, MEY drafted a second report, dated 5 March 1998, lifting its reserve. However, despite the introduction of the VAS, the fact that all or some of the pigs from certain batches may have ended up in cold stores means that a final reconciliation between the quantities delivered and discharged can only be made when all stocks in cold stores have been processed.

Cull sows

69. Cull sows were not initially eligible under the regulations that adopted exceptional support measures for the market in pigmeat in the four Member States. For the Netherlands, the basic regulation was amended(27) to introduce aid for up to 25000 cull sows at 70 % of the farmgate price for grade E pigs. Despite the fact that this measure was introduced at the request of the Dutch authorities it was never implemented by them, on the grounds that the welfare problem was not great enough to warrant buying-in. Although the protocols used for the "overname" scheme did not include the category of cull sows, 1000 of these animals were bought up under the scheme.

70. In Spain the basic regulation was also amended to introduce aid for up to 11000 cull sows, but only 50 % (5520) were bought up under the measure. Cull sows were not included in the amendments made to the basic regulations relating to Germany and Belgium.

Canning in Germany

71. Commission Regulation (EC) No 546/97 of 25 March 1997 amending Regulation (EC) No 414/97 adopting exceptional support measures for the market in Germany(28) authorised the use of pig carcases bought up under these measures for the production of canned meat. Given that there is always some risk that pigs may be infected with CSF, the meat must undergo heat treatment. The canning operation must be subjected to a permanent on-the-spot check of the meat processing by the authorities. To avoid disturbance on the Community market and to protect traditional trade links with third countries the canned meat must be exported outside the EU without refunds.

72. Compensation of DEM 8,2 million was paid for the purchase of 19392 pigs for the manufacture of canned products. 70 % (DEM 5,7 million) has been paid by the Community.

73. Two products were made: 1300 tonnes of "Schinkenfleisch" and 708 tonnes of "Schmalzfleisch", i.e. 50 % of the 4000 tonnes permitted under the Regulation. The entire production was destined for Russia. Exports to Russia began in week 20 of 1997 and ended in week 34 of 1998 as a result of the economic crisis in Russia. A total of 1594 tonnes was exported, but the remaining stock, 20 % of production, is in store at the production plant and must be subjected to customs controls to ensure that it does not enter the EU market.

The insemination ban in the Netherlands

74. Commission Regulation (EC) No 1564/97(29) provided aid for up to 220000 sows subject to the ban on insemination introduced from 3 June 1997 by the Dutch regulation Regulation banning pig breeding 1997 - Regeling fokverbod varkens 1997. The ban introduced on 3 June 1997 related to a designated area which was considerably extended on 24 June 1997. The aid is payable at ECU 32 per month per sow for the duration of the ban. Unlike the other market measures, the ban was compulsory. However producers could only claim the aid if they undertook to respect all the conditions. Initially in order to qualify for the aid producers had to ensure that the sows were inseminated within four months of the lifting of the ban. The Dutch authorities later realised that this would be impossible to enforce and the legislation was amended accordingly(30), so that the sows must be kept on the holding for the four months following the lifting of the ban, but need not be inseminated. The ban was lifted on 12 November 1997.

75. The total number of sows claimed under the Regulation was 191024. The Community contribution was limited to 70 % of the lesser of either the aid paid or the expenditure which would have been incurred in granting aid for the piglets that would have been born during the period of the ban (less 116 days) if the sows had been inseminated. The Dutch authorities have claimed ECU 9,3 million on the latter basis.

76. The insemination ban aid is subject to IACS controls. In fact, the Dutch authorities have inspected a far higher proportion (over 50 %) of claimants than the 10 % required. The results show a significant number of total rejections because the sows had been inseminated during the ban (34 out of 431 inspected). Given that the sample selection criteria included a risk assessment, these results should not be extrapolated.

Follow-up of the 1995 Annual Report

77. In its 1995 Annual Report the Court reported on measures taken by the German and Belgian authorities to combat the 1993 to 1995 CSF epidemic. The Court made four main recommendations (see paragraphs 28 to 31) which have not yet been fully addressed.

78. The legislation "should clearly state the conditions under which veterinary and exceptional market support measures can be implemented." Even if the Commission has meanwhile made clear its position with regard to the non-eligibility of certain categories of expenditure, some Member States persist in their claims for reimbursement. (see paragraphs 42, 50 and 63 to 66).

79. "The Commission should examine the appropriateness of different rates of support for the two measures." Exceptional market support measures have been financed at 70 % and veterinary measures at 50 %. Given the causal link between veterinary and market-support measures and difficulties experienced with classifying some measures, it would be more appropriate to introduce a common rate of co-financing at 50 %. This would remove any incentive to "reclassify" schemes for financial advantage.

80. "Steps must be taken to ensure that the compensatory amounts paid by the Member States to the producers are calculated using more precise and homogeneous methods". The rates of aid paid under veterinary measures reflect the degree of discretion granted to valuers by Member States. There are significant differences of rates of aid (market and veterinary) both between Member States and within them. This situation is exacerbated by the practice of mixed-batch weighing, reliance on weight estimation and use of differing weight categories (see paragraphs 53 to 58). Compensation for gilts (see paragraph 61) and sows has varied widely and in the Netherlands led to a situation whereby compensation for the latter reflected a notional number of piglets (see paragraph 53). Whilst it may be acceptable for rates of compensation to differ by Member State to reflect national market prices, it is not logical that in some cases veterinary rates exceeded market measure rates for the same type and weight of animal and in other cases the opposite occurred (see paragraph 62).

81. The Commission should bring forward to the Council a review of the regulations in force. It should examine the extent to which intensive pig producers should contribute towards the measures to combat classical swine fever: the Commission has produced a report on the situation in the pigmeat sector(31) which focuses on structural support measures. The report notes that the environmental costs of intensive pigmeat production systems are not internalised. This also applies to the costs of combating diseases such as CSF but no mention is made of the possibility of making producers meet some of these costs. The Dutch authorities have nevertheless taken steps in this direction by introducing a levy (see paragraph 7).

Conclusion

82. The cost-benefit analysis which served as the economic justification for the non-vaccination policy should be reassessed given the very high cost of the 1997 to 1998 CSF outbreak and the potential use of marker vaccines (see paragraphs 16 to 19).

83. The EU animal movement notification system failed to provide vital information necessary to combat the spread of the disease (see paragraph 24). In Spain CSF initially went undetected, thereby allowing it to spread (see paragraph 44). Certain individuals/companies in Germany, Spain and Belgium may also have contributed towards the spread of the disease (see paragraph 45). Failure to recognise semen as a vector of CSF aggravated an already difficult situation in the Netherlands and had repercussions in Belgium (see paragraph 52) and the decision to suspend preventive clearing in the Netherlands increased the problem of disease control (see paragraph 48).

84. The use of pig carcases to make canned meat for export required a great deal of administrative control. One third of the fattened pigs bought up under market support in Germany were used in this way, but 20 % of the final product remains in store with no prospect of an outlet in the short term (see paragraphs 71 to 73). Such measures, create more problems than they solve and should not be permitted in future. Similarly the insemination ban has proved difficult to control, a large number of claims were found to be ineligible (see paragraph 76).

85. Since the 1993 to 1995 epidemic, progress has been slow in developing a specific strategy to tackle subsequent outbreaks of the disease. In the mean time the Community has had to face its worst ever epidemic of CSF. Emphasis on other problems in the veterinary sphere (establishment of the free internal market and BSE) meant that CSF contingency plans were not given the priority the 1997 to 1998 epidemic has demonstrated that they deserved. The Court urges the Commission to dedicate the resources necessary to complete the revision of Community legislation and policy in this area as quickly as possible. Member States must also assume their responsibility for effective implementation of Community legislation in order to reduce the risk of future outbreaks of CSF causing epidemics on such a scale.

This report was adopted by the Court of Auditors in Luxembourg at its meeting of 13 January 2000.

For the Court of Auditors

Jan O. Karlsson

President

(1) Agricultural situation in the EU 1996.

(2) Commission Regulation (EC) No 2042/98 of 25 September 1998 (OJ L 263, 25.9.1998, p. 12).

(3) "Health strategies to control swine infectious diseases: European experience", Commission working document 30 January 1996 (VI/1715/96).

(4) Report on the situation in the pigmeat sector in the EU with a view to possible changes to structural support measures, COM(98) 434 final, 27.7.1998.

(5) Commission Decisions 98/399/EC of 8 June 1998 (OJ L 176, 20.6.1998, p. 36) and 1999/39/EC of 21 December 1998 (repealing Decision 96/552/EC) (OJ L 11, 16.1.1999, p. 47) respectively.

(6) "Health strategies to control swine infectious diseases: European experience", Commission working document 30 January 1996 (VI/1715/96).

(7) OJ L 47, 21.2.1980, p. 11.

(8) Commission Decision 1999/246/EC of 30 March 1999 (OJ L 93, 8.4.1999, p. 24).

(9) EUR 5738e.

(10) OJ L 47, 21.2.1980, p. 11.

(11) "The use of marker vaccines in the control of infectious diseases in particular", CSF VI/8119/97, 17.9.1997.

(12) Council Decision 93/384/EEC of 14 June 1993, (OJ L 1668.7.1993, p. 34).

(13) OJ L 22418.8.1990, p. 29.

(14) Untersuchungen zu Vorkommen, Ausbreitung und Bekämpfung der KSP in Deutschland 1993 - 95 (undated), Hanover Reference Laboratory.

(15) Commission Regulation (EC) No 413/97 (OJ L 62, 4.3.1997, p. 6).

Commission Regulation (EC) No 414/97 (OJ L 62, 4.3.1997, p. 29).

Commission Regulation (EC) No 581/97 (OJ L 87, 2.4.1997, p. 11).

Commission Regulation (EC) No 913/97 (OJ L 131, 23.5.1997, p. 14).

(16) Commisision Regulation (EC) No 370/98 of 17 February 1998 (OJ L 47, 18.2.1998, p. 10).

(17) Commission Decision 97/122/EC of 14 February 1997 (OJ L 45, 15.2.1997, p. 48).

(18) OJ L 224, 18.8.1990, p. 19.

(19) Commission Decision 1999/18/EC of 22 December 1998 (OJ L 6, 12.1.1999, p. 18).

(20) OJ L 16, 21.1.1998, p. 43.

(21) Commission Regulation (EC) No 76/97 of 17 January 1997 (OJ L 16, 18.1.1997, p. 74).

(22) Commission Regulation (EC) No 151/96 of 29 January 1996 (OJ L 23, 30.1.1996, p. 9).

(23) Information from Community CSF Reference Laboratory.

(24) Annex 8 to Final Evaluation Report by LNV (March 1998).

(25) OJ L 62, 4.3.1997, p. 26.

(26) Experiments carried out at the Reference Laboratory in Hanover at the beginning of 1998 suggest that the presence of the virus in semen is very difficult to detect and is only found in very low quantities. Because domestic boars do not manifest symptoms that are obviously indicative of CSF, all centres have tended to treat the symptoms rather than testing for CSF.

(27) Commission Regulation (EC) No 1293/97 of 3 July 1997, (OJ L 176, 4.7.1997, p. 23).

(28) OJ L 84, 26.3.1997, p. 12.

(29) OJ L 208, 2.8.1997, p. 25.

(30) Commission Regulation (EC) No 1688/97 of 29 August 1997 (OJ L 239, 30.8.1997, p. 1).

(31) Report on the situation in the pigmeat sector in the EU with a view to possible changes to structural support measures, COM(98) 434 final, 27.7.1998.

The Commission's replies

Summary

Whilst the Commission acknowledges that there is room for improvement, and that some measures have been delayed because of the demands made by more urgent priorities, it considers that the Court does not give due credit for the progress which has been achieved to date.

Although the Community has not been entirely free of classical swine fever (CSF), the evidence suggests that the most recent outbreaks are linked to the persistence of CSF in the wild boar population, and eradication of the disease in this group of animals presents particular difficulties, as noted in a recent report by the Scientific Committee.

The animal movement notification system Animo is undoubtedly capable of improvement, and the Commission will ensure that the Court's remarks are taken into account when the future of this system is reviewed. For the moment, the priority is to make sure that Animo continues to function after 1 January 2000. However, a substantial part of the criticisms that the Court lays at the door of Animo are attributable to the use made of the system by national officials rather than to shortcomings in the system itself. Some of them are, indeed, not even related to Animo, but to inadequacies in national administrative procedures. It should be remembered that Animo is essentially a cooperative venture between the Member States, and that the Commission plays the role of facilitator rather than of primary manager of the system.

The Commission agrees that its non-vaccination policy should be re-examined at the time that marker vaccines become reliable and available. However, a trial of these vaccines earlier this year showed that their development has not yet reached a stage where they can be used, and it is estimated that several more years' work remains to be done before they become a realistic option.

The Commission is currently financing a research project which is carrying out a cost-benefit analysis of alternative CSF control strategies, including the possible use of marker vaccines in the future. In the mean time, it remains essential for international trade reasons to maintain the non-vaccination policy.

As the Court notes, the Commission's inspections in the Member States have resulted in the reduction or refusal of requests for reimbursement of national expenditure on measures taken to deal with CSF outbreaks. The work on these matters is still in progress. The Court has participated in the Commission's inspections in one Member State in 1998 and there is no difference of opinion between the two institutions as to the ineligibility of the expenditure which was then examined.

It was the Commission which introduced the concept of Community and national ceilings on the number of animals eligible for exceptional market support measures, using the management powers at its disposal, during the 1993 to 1994 outbreak of CSF. Changes in the ceilings are entirely natural, as nobody knows at the outset how many animals will be affected by the outbreak. These decisions are taken by the Management Committee for Pigmeat, which is chaired by the Commission. Provisions are adopted after exchanges of views between the Commission and the Member States concerned, both during a preparatory phase and during the meetings themselves. However, the primary responsibility for carrying out these decisions lies with the Member States.

The Commission will propose that the co-financing rate for market support measures in future is the same as for veterinary measures, in order to avoid the difficulties described by the Court. It recalls, however, that for many years exceptional market support measures were financed at 100 %, and that it is on the Commission's initiative that the concept of co-financing has been gradually introduced.

The Commission shares the view that the producers should be made more responsible for the financial aspects of the eradication of animal diseases, e.g. through some form of insurance scheme. Since this matter falls within the competence of each Member State, the Commission intends, in the framework of the revision of Community legislation on these issues, to evaluate the possibility of supporting the Member States in establishing such schemes.

The Commission acknowledges that the work of ensuring that Member States submitted effective contingency plans for dealing with CSF has taken longer than it intended. There are several reasons for this. Firstly, the demands made on the Commission's human resources by the preparatory work for the introduction of the single market, at the same time as it became necessary to deal with the BSE crisis, meant that choices had to be made in prioritising the work. Since well-defined plans for dealing with outbreaks of foot-and-mouth disease were already in place, and could be used in the case of a CSF outbreak, it was considered that the preparation of CSF-specific plans could reasonably be delayed. Regrettably, the subsequent outbreak showed that the veterinary authorities of the Member States (who in any case have primary responsibility for disease control, in order to take the local epidemiological situation into account when controlling disease outbreaks) had underestimated the virulence and infectiousness of the CSF virus, and were not as well-prepared as they should have been.

The veterinary sector of the Commission has since been restructured and reinforced, both in the legislation and inspection fields, making possible an increase of activity related to the adoption of CSF contingency plans, taking into consideration the most recent problems encountered in the eradication of this disease. The Commission's Food and Veterinary Office, based in Dublin, is now engaged in a series of missions to assess the effectiveness of the Member States' contingency plans.

The Commission shares the Court's view that the 1997 to 1998 CSF epidemic has underlined the need to review some aspects of current legislation on this disease and its implementation, although it has been shown that CSF can be controlled and successfully eradicated if current legislation is properly applied. Taking into account experience and changes in scientific knowledge, however, the Commission will be making proposals for a revision of this Directive in the coming months.

Introduction

Principal features of the market

1 - 7. The Commission shares in principle the Court's description of the features of the market.

Regarding the structure of the pig industry, it should be noted that the absence of direct support measures or direct income aids for pig producers in the framework of the common agricultural policy has contributed to specialisation and concentration. Producers have been obliged to achieve economies of scale in order to earn a living and to maintain competitiveness.

Commission management

Eradication and surveillance programmes

13. Community initiatives aimed at more efficient prevention, control and eradication of CSF have been undertaken in recent years in all areas of the Community identifiable as at higher risk from CSF, in the framework of Council Directive 80/217/EEC and Council Decision 90/424/EEC.

Eradication plans are being implemented in all areas of Germany, France and Italy where CSF is endemic in the wild boar population. A surveillance programme is also being carried out on domestic pig holdings in Germany, in particular in areas at higher risk from CSF. In the rest of the Community, the CSF virus is not present in endemic form. If outbreaks occur, the contingency plans (see below) should ensure proper implementation of the measures established by legislation, and quick eradication of the disease.

Over the past two decades the Commission has made efforts to ensure that officially designated laboratories can diagnose CSF in an accurate and uniform manner in all Member States. In addition, research projects have been financed that have led to the development of molecular biology techniques able to distinguish rapidly between different strains of CSF virus. Epidemiologists thus have an important additional tool for tracing outbreaks of disease.

The Commission is well aware of the problems related to the presence of CSF in some central and east European countries and

Outbreaks of CSF in the European Union January 1997 to October 1999

(situation as at 25 October 1999)

>PIC FILE= "C_2000085EN.002301.EPS">

the future enlargement of the Community. A number of measures supported by the Commission under the Phare multicountry veterinary diagnosis and control programme concern better prevention and control of CSF, including:

- the training of scientists, official veterinarians and field veterinarians in diagnostic and border inspection procedures,

- the supply of laboratory equipment and establishment of common laboratory techniques,

- the preparation of contingency plans.

After the epidemic of 1997 to 1998 (the last outbreak occurred in Spain in July 1998), the incidence of outbreaks in the Community declined appreciably (see table). Even if the precise origin of the disease has not been fully clarified in some cases, the results of the epidemiological investigations carried out following the most recent outbreaks suggest a direct or indirect link with the persistence of CSF in the wild boar population. Unfortunately, the eradication of CSF in wild boar presents several difficulties, as noted in the report of 10 August 1999(1) by the Scientific Committee.

Contingency plans

14 - 15. The similarity of the control and eradication measures to be applied in cases of foot-and-mouth disease (FMD) and CSF is acknowledged by Community legislation. Article 14b of Council Directive 80/217/EEC states that the criteria to be applied mutatis mutandis for drawing up the contingency plans for CSF are those laid down for FMD.

Plans for the eradication of FMD were approved by the Commission in 1993 and 1995(2). Their provisions apply to a large extent also to CSF as they concern the resources required to deal with disease emergencies, and the persons involved in eradication measures are basically the same, i.e. farmers, the pigmeat industry, veterinarians, etc.

In view of the relevance of the analogous contingency plans for FMD, the Commission considered that the available human resources should be allocated in priority to activities related to the establishment of the free internal market(3), and to dealing with the BSE crisis. The veterinary sector of the Commission has since been restructured and reinforced, both in the legislation and inspection fields, making possible an increase of activity related to the adoption of CSF contingency plans, taking into consideration the most recent problems encountered in the eradication of this disease. However, this work required more time than expected, as some Member States delayed submitting plans and/or their later amendments.

The relevant Commission department (the Food and Veterinary Office in Dublin) has undertaken and is scheduled to undertake a number of missions to assess the potential effectiveness of the Member States' contingency plans, either specifically or as part of a disease control mission.

The Commission has reminded Member States on a number of occasions, during meetings of the Veterinary Committee, of the need to check the national plans for dealing with outbreaks of animal diseases.

In addition, the Animo system will be subject to more checks on its functioning and the implementation of relevant legislation by the Member States.

Vaccination

16 - 18. The non-vaccination policy adopted in the Community for CSF is based on eliminating CSF-infected herds. However, it also includes the possibility of vaccines being used in emergency situations, even if severe restrictions on trade must be applied in the areas where conventional vaccine is used. This is due to the impossibility of distinguishing between vaccinated and infected animals. This problem might be overcome if marker vaccines (and reliable laboratory tests for the detection of vaccinated but then infected animals) were available.

Following the advice of the Scientific Committee, the Commission financed a large-scale laboratory trial on the potential use of marker vaccines, which was carried out in eight laboratories in seven different Member States, and coordinated by the Community Reference Laboratory for CSF.

The trial came to an end in June 1999. The results indicated that marker vaccines would reduce the spread of infection within a herd and consequently reduce the risk of the disease spreading to other farms, but would not prevent transplacental transmission of virus in most of the vaccinated pregnant sows. In addition, the tests used to distinguish between vaccinated and infected animals are not suitable for use on individual animals and their sensitivity and specificity is inferior to the existing conventional tests.

The results of this trial will be taken into consideration by the Commission and the Member States, in the framework of the review of Council Directive 80/217/EEC (see paragraph 84).

19. Amongst the follow-up actions to the 1994 workshop referred to in the reply to paragraph 35 are the preparation and financing by the Commission:

- of a study on the epidemiology of CSF in densely-populated areas of Belgium and Germany (outbreaks of 1993 and 1995),

- beginning in mid-1998, of research project FAIR5-CT-3566 on the "Development of prevention and control strategies to address animal health and related problems in densely-populated livestock areas of the Community". This project is carrying out a cost-benefit analysis of alternative CSF control strategies, including the possible use of marker vaccines. The Landbouwuniversiteit, Wageningen (NL) will coordinate this project.

Notification of animal movements

24. The development of the Animo system has at all times been the responsibility of the Commission, which has taken all the necessary decisions to that end. Operating such a network at Community level requires a common host centre through which messages can be exchanged either directly between units in the system or through national host centres. Eurokom was designated as the common host centre for the computerised network following a call for tenders (see Decisions 91/638/EEC and 92/373/EEC). Arrangements for cooperation between the host centre and the Member States were laid down by Commission Decision 92/486/EEC as amended. The host centre should be regarded therefore as a provider of services to the Member States, which provide the funding. The Animo system was assessed at a seminar held in Portugal on 26 to 29 October 1997. During the seminar, a study carried out by Alcatel on the "Analysis and assessment of the current Animo network and system architecture" was considered. It was shown once again that there is a need for a common technical centre. The role of this centre is not to manage the system but to ensure that messages sent to it are forwarded.

26. The shortcomings of the Animo system in Spain have over a certain period been attributable to a variety of domestic causes. The issue has been discussed by the experts' group and in the Standing Veterinary Committee. From 1 November 1996, the national host centre in Spain was no longer capable of receiving messages from other Member States.

It should be noted that the Netherlands made a copy for the Spanish authorities of all messages concerning Dutch pig consignments. These copies were made available to the Spanish authorities from 10 February 1997.

The same procedure was also adopted for Italy, where two consignments of infected piglets from the Netherlands were traced and CSF quickly eradicated without any secondary outbreaks occurring. Consequently the statement by the Spanish authorities that the shortcomings of the Animo system were one of the main causes of the spread of CSF is at least debatable.

From April 1997, the management of the Animo system in Spain has been entrusted to Tragsatec. For a transitional period (April 1997 to January 1998), Tragsatec took responsibility for messages sent by the host centre using e-mail. It then distributed them to the various authorities concerned.

27 - 29. The Commission considers that improvements can be made to the present arrangements. This will probably mean involving users to a much greater extent rather than changing the whole system.

At the moment the Commission's priority is to see to it that the system is able to continue functioning after 1 January 2000, i.e. dealing with the problem facing every computerised system as regards both hardware and software. Once that date has passed, the Commission will take account of the Court's observations in order to improve the reliability of the system in the way arising from the seminar referred to above, and in the light of the legal deadlines under the present system.

Exceptional market support measures

31. The use of a ceiling for the number of eligible animals was introduced for the first time during the 1993 to 1994 epidemic. In order to provide for co-financing of the market support measures, ceilings were applied to both Community and national funding. Later on, the co-financing system followed the "70 % to 30 %" rule, but the principle of a ceiling was maintained. The ceilings were estimated by the Member States and adjusted regularly in line with the progress of the disease.

The ceilings can only be regarded as a guideline for the calculation of the financial aspects of the measure. They have to be adjusted from time to time to take account of outbreaks, and it is totally impossible to fix a precise number of eligible animals in advance.

The results of the Commission's on-the-spot inspection visits were taken into consideration when the support measures were drafted or amended.

On-the-spot controls by the Commission

34. The Netherlands case (in March 1997) was the first where the possibility of transmission of CSF through artificial insemination had to be considered seriously.

The matter is not specifically dealt with by Council Directive 80/217/EEC, but only by Directive 90/429/EEC on intra-Community trade and import of pig semen. However, the situation in the south of the Netherlands in the spring of 1997 was such that it should have led to a much more cautious approach to trade in semen from the restricted areas to the rest of the country.

The Commission is now fully aware of the risk and has given the matter its closest attention in a subsequent case during an on-the-spot check in the Federal Republic of Germany (late 1998) and in the review of Directive 80/217/EEC (see paragraph 84).

In the case of an outbreak of CSF, finished pigs cannot be sold on the market in the normal manner due to the veterinary and commercial restrictions. It is at this stage of the production process that market support measures have to be applied, offering farmers the possibility of delivering their fattening animals to the authorities. There is no need to buy in younger fattening animals as these may be marketed in the normal way if the epidemic ends early enough.

35. The Commission shares the Court's view that the 1997 to 1998 CSF epidemic has underlined the need to review some aspects of current legislation on this disease and its implementation, although it has been shown that CSF can be controlled and successfully eradicated also in high-density areas, if current legislation is properly applied.

However, this legislation includes a certain degree of flexibility that allows the competent authorities to control disease, taking into account the local epidemiological situation. It is up to the Member States to make proper use of this flexibility and if necessary to introduce further measures that go beyond the minimum requirements of Community Directives, when this is deemed necessary to achieve the objective, in accordance with Article 249 of the Treaty.

It is evident that eradication of certain animal diseases can involve additional difficulties in areas with a high density of livestock. The matter had been addressed several times in scientific meetings arranged by the Commission, well before the 1997 epidemic.

The Commission is financing further research on these issues (see also paragraphs 16 to 19) and the matter is also being dealt with in the framework of the ongoing review of Directive 80/217/EEC.

Community financing

37. The Commission wishes to underline that for many years exceptional market support measures were financed at 100 % by the Community budget. The principle of co-financing by the Member State was introduced by the Commission in stages in the ad hoc regulations opening exceptional market support measures related to the last two major epidemics (1993 to 1994 and 1997 to 1998). The first step was the introduction of the concept of Community and national ceilings on the number of animals eligible for the scheme. The second step was the application of the 70 % to 30 % sharing of the costs between the Commission and the Member State concerned.

The system of co-financing of market support measures was introduced by the Commission using the management powers at its disposal, because the Member States have the leading role in combating CSF. The engagement of the Member States' financial responsibility increased their awareness of the necessity of taking effective measures to combat the disease.

The Commission will propose that the co-financing rate for market support measures in future be the same as for veterinary measures, in order to avoid the difficulties mentioned by the Court.

Details of transactions under budget heading B2 - 5106, to which classical swine fever expenditure is charged, can be found in the Commission accounts using the Sincom system, to which the Court has access.

41. It is not yet possible, given the present situation regarding checks, in particular of expenditure for 1998 presented recently, to give precise figures for expenditure for Spain which will finally be recognised as eligible. It is likely, however, in view of the observations in principle on the finances that have already been notified officially to Spain, that a large part of the expenditure presented will not be recognised as eligible.

42. Preventive killings are one of the more effective ways of tackling the CSF problem in high-density areas such as Damme, Germany, and Wingene, Belgium, as the Court remarks in paragraph 46.

These slaughterings took place during the 1993 and 1994 epidemic and were financed as exceptional market support measures. The Court has already commented on this in its 1995 Annual Report, and as the Commission replied, the decision taken at the time was one of policy.

As indicated in the reply to paragraph 37, the alignment of the co-financing rate for market support measures with that for veterinary measures should eliminate the problem referred to by the Court at the end of the paragraph.

Implementation by Member States

Veterinary measures

Origin of the 1997 to 1998 epidemic

44. Issues relating to the Animo system have already been dealt with in the reply to paragraph 26.

Preventive killing

46. Council Directive 80/217/EEC and Council Decision 90/424/EEC contain provisions for preventive slaughter and destruction of pigs kept on farms which, due to the location of the farm or established epidemiological links to an infected holding, may be infected or contaminated with CSF virus. The objective of this is to reduce the possibility of the virus spreading further.

When CSF occurs in an area with a high density of pigs, the use of preventive slaughter is a useful eradication tool if the action is based on a thorough epidemiological investigation. Such an investigation must take into consideration all the potential factors that may have caused herd-to-herd virus transmission (quantity of virus spread in the environment before virus in the outbreak is removed, pig density in the area, distance to the outbreak, animal movements, direct and indirect contacts, etc.).

The local epidemiological situation will, in each case, determine the number and location of pig farms where preventive slaughter is to be carried out. The present legislation requires the Member States to perform in-depth epidemiological investigations and allows the use of preventive slaughter as required in a specific situation.

Unfortunately, some Member States underestimated the virulence and infectiousness of the CSF virus and were not adequately prepared to cope with the emergency situations which occurred in 1997.

47. The Court's report states that scientific opinion was divided at the time when preventive killing was first applied and then suspended in the Netherlands (February 1997). However, only a few Dutch experts opposed the measure. Unfortunately, their advice was followed. It must be underlined also that the Commission and the Standing Veterinary Committee were not informed in good time of the decision of the Dutch authorities to suspend preventive killing.

51. The Commission notified the Spanish authorities in September 1998 of its reservations regarding the eligibility of partial slaughter. Its position has remained unchanged.

Compensation

53 - 55. The Commission has drawn financial conclusions from the results of its checks and a large proportion of the expenditure for eradication presented by the Netherlands has been rejected. In particular, expenditure on the "overname" scheme has not been reimbursed (paragraph 64) and a financial correction of 25 % has been applied to the category of expenditure where irregularities have been found repeatedly. Together these corrections total some EUR 90 million.

57 - 61. The significant price inconsistencies and discrepancies referred to by the Court were in most cases mentioned also by the Commission in its inspection reports. In general these are the result of insufficient administrative supervision of compensation payments made by Member States at the time of the epidemic. The Commission will of course draw the appropriate financial lessons from its checks.

The Commission intends to issue guidelines to the Member States shortly, for the future but on the basis of experience gained during the 1997 epidemic, concerning both the procedures and the calculation of the compensation with regard to Community co-financing.

The Dutch "overname" scheme

65. Restrictions on the free movement of goods resulting from the application of veterinary measures are likely to bring about serious upheaval on the pigmeat market in the Member State concerned by an epidemic. This is the main reason for the introduction of exceptional market support measures. The normal market is closed, and delivery under the support scheme is the only possible outlet for the farmers concerned. Animal welfare aspects may also play a role but are not the principal reason for the support measure.

Exceptional market support measures

Purpose

66. Wider sampling and examination under the accounts clearance procedure showed the problem to be marginal and it will not be followed up in the clearance procedure.

Cull sows

69. Cull sows are replaced on a regular basis because their performance declines over time - 30 % to 40 % of sows are replaced each year. When an epidemic continues for several months, the farmer must be able to dispose of those animals in order to keep the economic results of his piglet production at an adequate level. It is up to each farmer whether he participates in the scheme or not. That is the reason also why cull sows have been eligible only in countries with a long-lasting epidemic (the Netherlands and Spain) but not in Germany or Belgium.

Canning in Germany

73. The 20 % of canned products still in store do not create costs for the authorities but for the company concerned. The canned products are stored under customs control and are clearly identified by their labelling in German and Russian.

The insemination ban in the Netherlands

74. The insemination ban was a new measure, introduced for the first time in summer 1997. During the first few months of the epidemic in the Netherlands, it was impossible to estimate its likely duration, and therefore the market support measures were limited to the killing of piglets and finished fattening pigs. Article 20 of Regulation (EEC) No 2759/75 states that market support measures "may be taken only in so far and for such period as is strictly necessary for the support of that market". This means that it is not possible, at the beginning of an epidemic, to introduce long-term measures. It only became clear during the summer of 1997 that the disease might continue until the end of the year, and that it would be reasonable to introduce an insemination ban, which would reduce the number of new piglets born four months later.

The insemination ban was applied for the first time at the specific request of the Dutch authorities. The Commission agrees that this instrument was not very successful and should not be repeated in the future, because the large-scale insemination of all sows at the end of the ban led to a disturbance of the piglet market in 1998.

76. In view of the need to respond to BSE epidemics and to undertake the work associated with the completion of the single market, as explained in the reply to paragraphs 14 to 15, the Commission had to make choices in prioritising its work in the animal health field. However, the work relating specifically to CSF has been reinforced since 1998.

Follow-up of the 1995 Annual Report

78. The Commission cannot prevent Member States from presenting ineligible claims for reimbursement of expenditure. It can, however, exclude those claims from reimbursement by the Community budget.

79. The Commission will propose that the co-financing rate for market support measures in future should be the same as for veterinary measures, in order to avoid the difficulties mentioned by the Court, as indicated in the reply to paragraph 37.

80. The Commission intends to issue guidelines to the Member States concerning both procedures and the calculation of compensation with regard to Community co-financing of veterinary measures, as indicated in the answer to paragraphs 57 to 61.

81. The Commission shares the view of the Court that producers should share greater responsibility for the financial aspects of the eradication of animal diseases. In the pig sector report, it was stated that farmers might contribute to insurance schemes for emergency situations due to animal diseases(4).

However, the whole matter falls under the competence of Member States. In some of them public or private/public insurance schemes have been established, which are also supported by pig producers.

It is very difficult from a technical point of view to create similar schemes at Community level. It is the intention of the Commission therefore to evaluate the possibility of supporting the Member States in establishing these schemes, in the framework of the revision of Community legislation on these issues.

Conclusion

82. A cost-benefit analysis of alternative CSF control strategies, including the possible use of marker vaccines, is currently being undertaken. As indicated in the reply to paragraph 19, the Commission is financing research project FAIR5-CT-3566 on the "Development of prevention and control strategies to address animal health and related problems in densely-populated livestock areas of the Community". The Landbouwuniversiteit of Wageningen (NL) is the coordinator of this project.

The principles of the non-vaccination policy are the same in all developed countries and any change in the EU policy must adequately consider the consequences on intra-Community and international trade.

By way of comparison, over the last decade the slaughter policy in the case of foot-and-mouth disease has produced direct savings on vaccination costs estimated at over ECU 1 billion. The indirect economic advantages related to increased trade within the Community and exports to third countries are very difficult to estimate but are likely to be many times higher. In the mean time the number of disease outbreaks in the Community and the related costs and losses have been much lower than in the previous 10 years.

83. The Commission will take account of the Court's comments in improving the reliability of the Animo system, as mentioned in the reply to paragraph 29. It does not however agree that a failure of the system was the principal cause of the failure to detect the spread of the disease in Spain (see reply to paragraph 26). As far as the failure to recognise semen as a vector of CSF is concerned, the Netherlands case was the first where the possibility of transmission of this disease through artificial insemination had seriously to be considered. However, the situation in the Netherlands in April 1997 was such that it should have led to a much more cautious approach on trade of semen from the restricted areas to the rest of the country. The Commission is now fully aware of the risk (see reply to paragraph 34). As far as the suspension of preventive killing in the Netherlands is concerned, the Commission recalls that neither it nor the Standing Veterinary Committee were involved in the discussion prior to the decision to suspend preventive killing.

84. The storage of the canned products is a matter for the company concerned only, as noted in the reply to paragraph 72. As far as the insemination ban is concerned, the Commission agrees that this instrument was not very successful and should not be repeated in the future (see reply to paragraph 73).

85. Since the 1993 to 1995 epidemic the Commission has:

- seen to it that all Member States have a contingency plan for FMD, which is the most economically important viral disease in pigs, and that these plans are available for use in the case of an outbreak of CSF,

- requested and obtained an opinion from the Scientific Veterinary Committee on the potential advantages and disadvantages of the use of a marker vaccine during a disease epidemic,

- conducted a large-scale laboratory trial on the use of marker vaccines, for which marketing authorisation has been sought but not yet obtained,

- supported research project FAIR5-CT-3566, referred to in the reply to paragraph 81, which includes a cost-benefit analysis of alternative CSF control strategies.

The CSF epidemic in 1997 to 1998 underlined the need to review some aspects of current legislation on this disease and its implementation. In December 1998, therefore, the Commission together with the Member States' experts launched a review of Community legislation on CSF (Council Directive 80/217/EEC).

This review will consider experience gained in implementing disease eradication measures, scientific progress in the fields of CSF diagnosis and the development of marker vaccines, problems relating to persistence of disease in the wild boar population, the results of most recent studies on CSF epidemiology, cost-benefit analysis of different eradication strategies, and changes in the structure of the pig industry. No basic changes are foreseen, however, in the non-vaccination policy adopted in the Community, which is supported by all Member States.

A Commission proposal for the amendment of Directive 80/217/EEC will be sent to the Council in the coming months. It should continue to recognise the need for a certain degree of flexibility to allow the competent authorities in the Member States - who in any case have primary responsibility for disease control - to take the local epidemiological situation into account when controlling disease outbreaks.

(1) Document XXIV/B3/R09/1999 is available on the Internet at the following address: http://europa.eu.int/comm/dg24/health/sc/scah/out24_en.pdf

(2) Commission Decisions 93/455/EEC and 95/194/EC.

(3) In the early 1990s a lot of work was done by the Commission and the Council to legislate on veterinary matters in order to ensure the establishment of the free internal market. This involved the adoption of about 90 Directives. The Commission also had to coordinate the efforts of the Member States to implement this legislation.

(4) Report on the situation in the pigmeat sector in the EU with a view to possible changes to structural support measures, COM(98) 434 final, 27 July 1998, p. 25.