Court of Justice 17-10-2002 ECLI:EU:C:2002:595

Is it possible to apply the de minimis principle?

27. The Danish Government argues, however, that the problem is extremely limited, since out of a total of 1 300 bathing resorts results which failed to meet the standards were recorded in only 130 resorts during the period 1995-1998, and of those 130 only 8 resorts recorded more than one deviation during that period, namely six resorts on two years and two resorts on three years.

28. The defendant adds that by requiring compliance with the standards in all bathing areas the Commission makes it impossible in practice to comply with the directive and requires Member States to impose bathing bans in situations in which neither public health requirements nor environmental protection requirements justify them.

29. It is necessary to take into account the fact that the objective of the directive is also to preserve the ‘quantitative quality of bathing water’. That would be jeopardised if the directive were interpreted in such a way as to force Member States to reduce considerably the quantity of bathing waters accessible to the population through unjustified banning measures.

30. The directive should therefore be interpreted in the light of the de minimis principle. The defendant by no means denies that there is no general de minimis principle within the area of secondary legislation and is not seeking a declaration of such a general principle from the Court. It considers however that in the specific case of a directive which it would be impossible to implement without the de minimis rule it is necessary to apply that principle.

31. The Danish Government recognises in that connection that under Article 5(1) of the directive it is sufficient that in a given bathing area 95% of samples comply with the parameters. However, if the period during which the samples are taken is five months, as in Denmark, and the Member State takes two samples per month, that is to say, 10 samples in total, as stipulated by the directive, it is sufficient for a single sample not to comply with the standards for the resort to be regarded as being polluted. In that case, the rate of compliance falls to 90% instead of the 95% allowed by Article 5.

32. If it were not possible to apply the de minimis rule the Commission's approach would in fact amount to requiring 100% compliance for each bathing area and thus render Article 5(1) redundant.

33. What is to be thought of this view?

34. There is no doubt that the Danish Government's reasoning is correct. However, as the Court noted in paragraph 36 of Commission v Germany, cited above, the directive only specifies the minimum frequency of sample-taking and does not therefore preclude the Member States from increasing the number of samples, thus reducing the proportion represented by samples not satisfying the stipulated conditions. Thus, if 20 samples are taken a Member State may ‘allow itself’ one sample that does not comply (5% of 20 = 1). If 30 samples are taken the percentage of 5% is equivalent to 1.5 samples. This means that even in that case only one sample that does not comply is tolerated. As it is hardly conceivable that a Member State could take more than 30 samples in the same bathing area the result of the directive is in practice that having found that two samples do not comply with the standards the Member State has no need to continue taking samples: it only remains for it to prohibit bathing in the area in question.

35. The regulations are therefore very strict, although it is important not to lose sight of the fact that for certain substances, such as ‘faecal coliform’, the margin laid down by Article 5 is 20%. At any event, it is undeniably clear from the case-law, and in particular the judgment cited above, that the obligations laid down in the directive are indeed as I have just described.

36. In particular, it is clear that the Court did not intend to allow room for the application of any de minimis principle within the framework of the interpretation of the directive.

37. It is clear from the foregoing that the infringement concerning bathing water quality claimed by the Commission is established, since the defendant does not deny that during 1995 to 1998 the water in a certain number of areas did not reach the standards provided for in the directive, as its own figures show.

38. It is solely for the purposes of completeness therefore that I shall consider the remainder of the parties' arguments.

It is apparent that the differences between the figures submitted by the two parties result from divergent understandings of the scope of the obligations imposed by the directive.

Is it necessary to assess the state of the water every year?

39. First, the defendant challenges the way in which the Commission acted, on the grounds that it is necessary to assess the cleanliness of bathing water over several years and not on the basis of figures which are valid for only one year. Such a method of calculation serves to give a distorted statistical picture of the quality of bathing water in Denmark and is by no means supported by the directive.

40. The Commission contends in this regard that it is expressly stated in Article 13 of the directive that it is necessary to check each calendar year whether the requirements of the directive are being complied with. Furthermore, and at any event, the quality of bathing water still does not comply with the directive in 1999 and 2000, which proves the inadequacy of the measures taken by the Danish authorities. Lastly, the Commission adds that it only brings an infringement action where the infringements recorded continue over several years.

41. It is undeniable that the fact that the directive requires Member States to send the Commission a report each year on the implementation of the directive for the current year proves that it is necessary to assess the situation with regard to bathing water every year. That conclusion is confirmed, moreover, by the objective of the directive. Protection of public health is at issue in the present case and that consideration cannot accommodate long periods during which the authorities of the States are entitled to wait before taking action. In particular, it would be unacceptable for several bathing seasons to pass before a Member State, finally convinced that a bathing area did not comply with the directive, took the necessary measures.

42. Consideration of the case-law leads, moreover to the same conclusion since, in paragraph 34 of Commission v Germany, cited above, the Court expressly held that if values were exceeded only once in only one season that constituted an infringement of the directive.

Is it possible to exclude accidental deviations?

43. Second, Denmark disputes the Commission's figures on the grounds that they also take into account accidental deviations.

44. In that regard the defendant argues that except for 9 cases all of the deviations from the thresholds laid down by the directive which the Commission took into account are due to unpredictable natural causes. It mentions bird excreta in particular: this may be concentrated in a few litres of bathing water sampled in a given area and occurs suddenly and unpredictably. It is appropriate to stress in particular that such phenomena do not occur regularly in the same areas but are distributed differently each year.

45. Denmark does not rebut the Commission's statement that it is necessary to show caution when designating bathing areas in regions where there is an abundance of wildlife. It states, however, that even such a cautious approach does not necessarily have the effect of reducing the number of accidental deviations.

46. It therefore considers that the Commission should have taken into account the reasons for the deviations and excluded from its table accidental deviations caused by birds and other animals, which do not point to a general deterioration in bathing water quality and which cannot be prevented.

47. The Commission quotes the settled case-law of the Court,⁽⁵⁾ which states that the directive requires Member States to take all necessary measures to ensure that bathing water conforms to the limit values laid down in the directive and does not allow them merely to take all reasonable measures possible.

48. In that regard, Denmark does not deny it has an obligation as to the result to be achieved but it contends that that obligation is not absolute. It is entitled to rely on the absolute impossibility of complying with the provisions of the directive.

49. The Commission, however, contends that the directive contains no basis for making adjustments for accidental deviations such as those described by the Danish Government in order to discount deviations due to birds or other animals.

50. In the Commission's view, it is hard to consider pollution due to birds and wildlife in general as a fortuitous event and it should therefore be taken into account in the forecasts. Member States may, in that regard, increase the number of samples in order to reduce the proportion represented by samples which do not comply due to fortuitous circumstances.

51. I share that view.

52. The directive does, however, provide for a certain number of situations in which it is possible to discount samples revealing deviations.

53. The directive provides in Article 8, cited above, that it may be waived under exceptional weather or geographical conditions or when bathing water undergoes natural enrichment in certain substances coming from the soil causing a deviation from the values prescribed in the annex. Moreover, Article 5(2) provides that deviations are not to be taken into consideration when they are the result of floods, other natural disasters or abnormal weather conditions.

54. It is hardly possible to argue that those provisions are applicable to animal excreta, nor does the defendant claim they are.

55. The directive also provides for the possibility of a deviation in the normal course of events, that is to say, without stipulating any exceptional circumstances. It should be pointed out that Article 5(1) of the directive provides that samples are to be deemed to conform to the requirements of the directive where a certain percentage of the samples, ranging from 95% down to 80% in the case of ‘faecal coliform’, meet the values laid down in the annex to the directive.

56. It is apparent therefore that the Community legislature took into account the need to make provision for a certain margin for deviations which it regarded as not representing a structural deterioration in water quality.

57. The defendant's arguments, however, raise the question whether it is appropriate to include in the interpretation or application of the directive a further margin in order to take into account the impact of animal excreta.

58. These cannot be described as exceptional circumstances and, one is tempted to say, are more in the natural course of events. It is difficult to see why, on the face of it, samples which are non-representative on those grounds should not come into the margin provided for in Article 5(1) of the directive. There is no discernible reason why they should be treated differently from any other potential sources of distorted results.

59. It must be repeated in this context that the event in question is by no means extraordinary. It follows, first, that it is hardly conceivable that the Council, when it adopted the directive, and in particular Article 5 of the directive, was unaware of the possible implications of that phenomenon.

60. It is true, as I have shown above, thai the tolerance threshold is generally 95%. which allows in reality only a single sample that does not comply. The question arises, therefore, whether as part of the revision ol the directive, which was mentioned at the hearing, the Community institutions should not consider raising that margin of tolerance in cases where there is no other source of pollution apart from bird excreta. That will depend perhaps on the view one takes of the significance of the dangers to the health of bathers represented by such excreta.

61. Be that as it may, the phenomenon in question cannot be classified as a case of force majeure, which according to the case-law⁽⁶⁾ is a circumstance which a Member State can plead to justify its failure to comply with its obligations.

62. The fact that samples of bathing water are distorted by animal excreta does not constitute an unpredictable and insurmountable obstacle to compliance with the rules laid down in the directive.

63. It is appropriate to point out in that regard that the defendant itself observes that pollution of animal origin is often detected in areas with particular characteristics, such as shallow water or a lower level of water circulation. The authorities are therefore in a position to determine the areas at risk of that type of pollution and to pay them particular attention without having to resort to a disproportionate number of samples. The Danish Government itself states that in such areas it automatically takes 20 samples a season.

64. Measures can therefore be adopted, once distorted samples have been taken, to prevent the Member State concerned from finding itself in breach of the directive.

65. The defendant however expresses various criticisms of the remedy of increasing the number of samples. It contends that that method cannot be effective because it involves high costs and is out of proportion to the results to be expected from it.

66. According to settled case-law,⁽⁷⁾ such arguments do not, however, justify failure on the part of a Member State to comply with the provisions of a directive.

67. The defendant also states that even where the number of samples complying need only be 90% or 80% an increase in the number of samples will not have any impact on water quality and any improvement recorded will be only statistical.

68. Although it is indisputable that the simple fact of increasing the frequency of sampling will not in itself improve water quality, it should be pointed out that, as we have seen, the increase makes it possible to determine the permanence or frequency of the pollution and thus enables the competent authorities to take the necessary decisions more quickly if the pollution has become too frequent to be tolerated.

69. In that case the national authorities should impose bathing bans in the affected areas if pollution caused exclusively by animals persists and it is therefore not possible to take action with regard to the cause of the pollution.

70. The defendant explains in that regard that that is exactly what it did in a certain number of cases. It infers from this that areas which have been the subject of a ban should not be included among those where the requirements of the directive are not met.

71. The Commission disputes that reasoning, arguing that a Member State may not evade its obligations under the directive by imposing a temporary ban on bathing because the bathing water does not comply with the requirements of the directive. If a Member State wants a bathing area not to be subject to the obligations under the directive the exclusion should be definitive and permanent.

Does the breach continue even if a bathing ban is imposed?

72. It is true that the Commission's view appears to follow the case-law of the Court in so far as we read in paragraph 33 of Commission v Germany, cited above, that the fact that areas have lost their status as bathing areas or that measures have been taken to remedy the infringement does not cure it.

73. However, as the Danish Government rightly observes, ‘it is physically impossible to, react before a deviation from the limit values is recorded or the sources of the pollution identified. The deciding factor must therefore be whether Member States take all the necessary measures once the deviation is established’.

74. It follows that where a Member State has taken the only measures it is possible for it to take, in this case a ban on bathing, there is no reason to find that it has failed to fulfil its obligations.

75. Moreover, and above all, Article 1 of the directive expressly provides that the directive applies to bathing water, that is to say water in which bathing ‘is explicitly authorised by the competent authorities of each Member State or is not prohibited and is traditionally practised by a large number of bathers’. It could not be stated more clearly that it does not apply to water in which bathing is prohibited.

76. The conclusion to be drawn therefore is that where a Member State bans bathing in a specified area, even if it is only during the season, that area is not to be taken into consideration for the purposes of assessing implementation of the directive in that Member State.⁽⁸⁾

77. That conclusion is, moreover, borne out by the objective of the directive. The directive seeks to ensure that Member States take adequate measures to protect the health of bathers, where it may be regarded as being put at risk, and not to penalise them even in situations where they have taken protective measures immediately the dangerous nature of the situation became apparent to them.

78. I do not find anything either in the directive or in case-law to support the Commission's view that only a definitive ban will meet the requirements of the directive. The directive does not draw any distinction between bans, whether definitive or temporary. Moreover, the prospect of the possibility of reopening is likely to encourage the competent authorities to take the necessary measures to ensure the area complies with the requirements. The rule that any ban should be permanent, however, would be a disincentive for the authorities to seek to remedy the deviations that had been established. The Commission is careful to point out, moreover, in what circumstances which are relevant as regards the objectives of the directive such drastic bans should be imposed.

79. Lastly, the argument the Commission seeks to draw from Commission v Belgium is unconvincing.⁽⁹⁾ In that case the Court ruled that the fact that the number of bathers falls below a certain level in a given area does not permit a Member State to consider that the area falls outside the scope of the directive. The Commission is wrong to infer from this that the fact that bathing is prohibited in an area does not mean that the area no longer falls within the provisions of the directive.

80. Unlike a factual situation in which the number of bathers has fallen, a ban on bathing implies necessarily that the objective of protecting public health, the importance of which was noted by the Court in that case, is no longer at issue since, by definition, no bathers are exposed to any risk