Petsafe Ltd & Anor, R (on the application of) v The Welsh Ministers [2010] EWHC 2908 (Admin)

[2010] EWHC 2908 (Admin)Case No CO/5875/2010
IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT
Venue Cardiff Civil Justice Centre, 2 Park Street, Cardiff, CF10 1ETDate 16/11/2010
Before
THE HONOURABLE MR JUSTICE BEATSON
The Queen on the application of
Between
Petsafe LtdFirst ClaimantThe Electronic Collar Manufacturers AssociationClaimantThe Welsh MinistersDefendant
(Transcript of the Handed Down Judgment ofWordWave International LimitedA Merrill Communications Company165 Fleet Street, London EC4A 2DYTel No: 020 7404 1400, Fax No: 020 7404 1424Official Shorthand Writers to the Court)
Monica Carss-Frisk QC and Hanif Mussa (instructed by Brabners Chaffe Street LLP) for ClaimantsClive Lewis QC and Joanne Clement (instructed by Welsh Assembly Government) for DefendantHearing Hearing date: 8 October 2010
Judgment

Introduction

[1]The issue in this case is whether the Welsh Ministers were entitled to prohibit the use on cats and dogs of any electronic collar designed to administer an electric shock as they did by the Animal Welfare (Electronic Collars (Wales)) Regulations 2010 (SI 2010/934) (“the 2010 Regulations”). The 2010 Regulations are made under section 12 of the Animal Welfare Act 2006 (“the 2006 Act”). Section 12 (1) empowers “the appropriate authority”, in Wales the Welsh Ministers, to make regulations “for the purpose of promoting the welfare of animals for which a person is responsible, or the progeny of such animals”. By the 2010 Regulations using such a collar is an offence punishable with up to 51 weeks imprisonment.[2]The first claimant is a manufacturer and distributor of pet products. The products it distributes include electronic collars. The second claimant is an unincorporated association of four companies, including the first claimant, which manufacture and distribute electronic training aids for animals. They seek permission to apply for judicial review of the 2010 Regulations and, if permission is granted, an order quashing the Regulations, which they submit are invalid. A previous challenge (CO/9043/2008) was launched in September 2008. Permission in that case was refused on 31 March 2009 on the ground of prematurity.[3]The use of electronic collars and similar devices is controversial. A number of groups, including the Kennel Club, have been campaigning for some time to ban them because they have the potential to have adverse consequences for animals, and are cruel and unnecessary. Others maintain that the scientific evidence does not support a ban or regulation, and that the devices help to avoid injury to animals on roads or at the hands of farmers protecting their stock. During 2007 a number of governmental bodies in the United Kingdom considered the question. In England, the Department for Environment, Food and Rural Affairs (“DEFRA”) commissioned research to assess the effect of pet training aids on the welfare of domestic dogs. In Scotland and Wales there were consultation exercises to consider whether electronic collars are harmful to dogs and cats and whether they should be banned.[4]The Welsh Ministers’ first consultation ran from November 2007 to February 2008. In the light of the responses the Welsh Ministers decided to ban electronic collars. There was a further consultation between 4 March 2009 and 25 May as to the terms of draft regulations to do this. That consultation proposed two defences or exemptions. The first was where the collar was attached by or under the direction of a veterinary surgeon. The second was where the device was used for a short period and was activated by a boundary marked by a physical barrier. The second consultation was delayed because of the earlier judicial review. There was a third consultation between 21 December 2009 and 8 February 2010 in the light of the responses to the second consultation. In particular the two defences canvassed in the second consultation were removed.[5]The outcome of the consultation process was summarised by the Office of the Chief Veterinary Officer for Wales (“OCVO”), a division of the Department for Rural Affairs of the Welsh Assembly Government in an Explanatory Memorandum. The summary and a draft of proposed regulations were laid before the National Assembly for Wales on 14 February 2010. The 2010 Regulations were made on 23 March and came into force on 24 March. These proceedings were launched on 21 May 2010. The application for permission was accompanied by an application for urgent consideration on the ground that there was a risk of wrongful prosecution and conviction. The papers came before His Honour Judge Jarman QC. He considered that, while it was not clear that the claimants had demonstrated reasonable arguments that the 2010 Regulations constitute an unjustifiable or disproportionate means of promoting animal welfare, given the importance and potentially wide impact of the issues, the court would be assisted by more detailed submissions as well as by oral argument.[6]The claimants’ original claim form submitted that the 2010 Regulations violate the European Convention on Human Rights and EU Law. It is contended that the decision to ban electronic collars was disproportionate to the aim of promoting animal welfare. It is said the Regulations are thus contrary to Article 1 of the First Protocol to the European Convention on Human Rights and the prohibitions in Articles 34 and 56 of the Treaty on the Functioning of the European Union. Article 34 prohibits unjustified restrictions on the free movement of goods. Article 56 prohibits unjustified restriction on the freedom to provide services. Secondly, the 2010 Regulations are challenged on domestic public law grounds. The grounds are irrationality, Wednesbury unreasonableness, and perversity. At the hearing Miss Carss-Frisk QC abandoned the Article 56 point. She also accepted that, as far as the proportionality of the 2010 Regulations is concerned, the claimants’ case based on Article 34 is stronger than that based on Article 1 of the First Protocol.[7]Three days before the hearing, in her Skeleton Argument, Ms Carss-Frisk QC raised two new domestic law grounds. The first is that the acts which are criminalised by regulations 2(1)(a) and (b) are not qualified or restricted in their application to animals “for which a person is responsible” as required by section 12 of the 2006 Act. She submitted that they are thus outwith the power conferred by it. The second concerns regulation 2(1)(c). Although that provision creates a prohibition by reference to the statutory concept of responsibility for an animal, it is submitted that it has perverse consequences. This is because it makes it a criminal offence for a person to be responsible for a cat or a dog to which an electronic collar is attached irrespective of the reason for which the person is taking responsibility for that animal. It thus includes responsibility assumed by a person in order to remove the electronic collar or otherwise to care for the welfare of the animal. During the hearing amended grounds of challenge, dated 7 October, were provided to the court and the defendant.

The Evidence

[8]The evidence in support of the application consists of the statements of Ms Angela Critchley, the first claimant’s International Marketing Manager, dated 19 May 2010, Mr David Chamberlain, veterinary consultant to the first claimant, dated 20 August 2010, and the two statements, in almost identical terms, of Lord McNair, respectively dated 20 August and 13 September 2010. Lord McNair is the national spokesman for the second claimant. His statement dated 13 September 2010 does not identify the changes from his earlier statement or give any explanation for them. It contains no indication that it is a second statement. At the hearing, Miss Carss-Frisk QC stated that the changes were either corrections or further particulars of matters in the first statement. That may be so, but it is surprising that a statement prepared with the assistance of solicitors should be prepared in this way. A statement of Lord McNair dated 23 September 2008, prepared for the earlier judicial review was also before me. It contained both factual evidence and the grounds, submissions and authorities relied on in those proceedings.[9]The evidence on behalf of the defendant consists of two statements of Professor Christianne Glossop, the Chief Veterinary Officer for Wales and Head of OCVO, and one of Mr Huw Jones, OCVO’s Head of Animal Welfare. Professor Glossop’s statements are dated 12 July and 20 September 2010. Mr Jones’s statement is dated 17 September 2010.[10]On 20 August the claimant applied for permission to rely on three undated and unsigned statements made by witnesses identified only as Mr X, a resident of Wales, Ms Y, a resident of Monmouth, and Mr Z, a resident of Wrexham. These witnesses stated that they desired anonymity because of concerns for the welfare of their family and themselves in the light of “the sensitive nature of these issues and the well-known reprisals inflicted by animal rights activists upon those willing to support anything which they oppose”. The claimants were willing to disclose their names to the court.[11]There were a number of difficulties with this application. The names and addresses of virtually all the respondents to the defendant’s consultations had been published. There was, however, no evidence before the court of any threat from animal rights protestors in relation to those persons, these proceedings, the question of electronic collars, or these witnesses. Secondly, the claimants did not respond to a request dated 17 September by the defendant for disclosure to it of the names of these witnesses on a confidential basis. It was only after the commencement of the hearing and after being pressed that they agreed to disclose the names of Mr X and Mr Z to the defendant on this basis. Ms Y did not consent to disclosure on this basis because she feared a leak. It was difficult to see how the defendant could, at that late stage and after the commencement of what was to be a one day hearing, check on the reliability of the witnesses or deal with their evidence. In these circumstances I concluded that it would be unfair to the defendant to admit the statements and that, if admitted, little if any weight could be given to them. I also considered that it was inappropriate for a party to seek to adduce the evidence of anonymous witnesses without an adequate evidential basis for doing so. I do not consider that the subjective fears of the makers of the statements, unsupported by any objective evidence, suffice.

The Legislative Framework

[12]The Animal Welfare Act 2006 provides for a series of substantive criminal offences on those who inflict harm on animals and on those responsible for an animal who fail to take steps to ensure its welfare. It also empowers public authorities to create further obligations to promote animal welfare by regulations, licensing or registration of activities, and codes of practice.[13]The 2006 Act in general excludes invertebrates. It distinguishes the following three categories:
“protected animals”, animals that are owned or for which a person is otherwise responsible, and all animals within the scope of the Act.

(c) it is not living in a wild state”

. Stray dogs and feral cats thus fall within (a), but animals of a kind not commonly domesticated in the British Isles are only “protected animals” to the extent that they are under the control of man, or not living independently in the wild.[14]Section 3 of the 2006 Act defines “responsibility for animals”. It provides:
“(1) In this Act, references to a person responsible are to a person responsible for an animal whether on a permanent or temporary basis. (2) In this Act, references to being responsible for an animal include being in charge of it. (3) For the purposes of this Act, a person who owns an animal shall always be regarded as being a person who is responsible for it. (4) For the purposes of this Act, a person shall be treated as responsible for any animal for which a person under the age of 16 years of whom he has actual care and control is responsible.”
Section 62(4) provides that references to responsibility in the 2006 Act are to be read in accordance with section 3.[15]The offences of causing unnecessary suffering to an animal (section 4), mutilation (section 5) and administration of poisons (section 7), only apply to “protected animals”. That concerning the docking of dogs’ tails (section 6) applies to all dogs except certain certified working dogs. That concerning “fighting” (section 8) applies to all animals within the scope of the 2006 Act. An obligation to ensure the welfare of animals is imposed on those who are responsible for the animal by section 9 of the Act.[16]Sections 9 to 12 of the 2006 Act contain provisions dealing with the promotion of welfare. Section 12 provides:
“12 regulations to promote welfare (1) The appropriate national authority may by regulations make such provision as the authority thinks fit for the purpose of promoting the welfare of animals for which a person is responsible, or the progeny of such animals.”
Section 12(3) provides that the power to make regulations includes the power to provide that the breach of a provision of the regulations is an offence, and to apply a relevant post-conviction power in relation to conviction for an offence under the regulations. Section 12(5) provides that a specified offence under the regulations is to be treated as a relevant offence for the purpose of the power in section 23 to issue a warrant authorising premises to be entered and searched.[17]The Welsh Ministers are the authority with power to make regulations in relation to Wales: see the effect of section 162, and paragraph 30 of Schedule 11 to the Government of Wales Act 2006 (“GOWA 2006”) on the definition of “appropriate national authority” in section 62(1) of the 2006 Act. By section 61, regulations made by the Welsh Ministers require approval by an affirmative resolution of the National Assembly for Wales.[18]I have stated that the 2010 Regulations were made on 23 March 2010 and came into force on 24 March. The preamble to the Regulations states that the Welsh Ministers made them, exercising the powers conferred on them by sections 12(1), (3)(a), (5) and 62 of the 2006 Act, and having consulted such persons appearing to them to represent the interests concerned. The Regulations prohibit the use of any electronic collar which regulation 1(2) defines as “a collar designed to administer an electric shock”. regulation 2 provides:
“2. Prohibition on use of electronic collars (1) It is prohibited for a person to – (a) attach an electronic collar to a cat or a dog; (b) cause an electronic collar to be attached to a cat or a dog; or (a) attach an electronic collar to a cat or a dog; (b) cause an electronic collar to be attached to a cat or a dog; or (c) be responsible for a cat or a dog to which an electronic collar is attached.”
[19]By regulation 3 a person who breaches any of the prohibitions in regulation 2 commits an offence which is summarily triable and is liable to imprisonment for a period not exceeding 51 weeks and/or a fine not exceeding Level 5 on the standard scale. Other types of electronic device which are used to control animals including what are known as “scat-mats” are not covered by the 2010 Regulations.[20]I have referred to the power to issue codes of practice. Between 1 December 2008 and 30 October 2009 the Welsh Ministers issued four codes of practice dealing with the welfare of respectively cats, dogs, equines, and rabbits. Only the text of the code for dogs was before me. It is common ground that each code contains sections on protecting an animal from pain, suffering, injury and disease. It is stated in paragraph 13 of the claimants’ grounds that none makes any mention of electronic collars or to the risk of harm to an animal which wears one. The Code of Practice for the Welfare of Dogs, which came into force in November 2008, states a preference for positive training methods using rewards for good behaviour rather than punishment: see especially paragraph 3.8.[21]Article 34 of the Treaty on the Functioning of the EU provides that “quantitative restrictions on imports and all measures having equivalent effect shall be prohibited between Member States.” By Article 36, Article 34 does not preclude prohibitions or restrictions which are justified inter alia on the ground of the protection of health and life of animals. Article 1 of the First Protocol to the European Convention on Human Rights provides that “every natural or legal person is entitled to the peaceful enjoyment of his possessions” and that no-one shall be deprived of them “except in the public interest and subject to the conditions provided for by law”.[22]The 2010 Regulations and the preceding draft regulations which were published on 4 March and 21 December 2009 are subject to the notification and “stand-still” provisions of Directive 98/34/EC. Article 8 of that Directive requires Member States to “immediately communicate to the [European] Commission any draft technical regulations” except where the draft merely transposes the full text of an international or European standard. The Commission is required immediately to notify other Member States of the draft. The Commission and other Member States are empowered to make comments to the Member State which has produced the draft regulation and that State is required to take such comments into account when enacting the regulations. Member States are also required to communicate the definitive text of the regulations to the Commission. Article 8 also provides that the absence of a reaction from the Commission under the Directive to a draft technical regulation “shall not prejudice any decision which might be taken under other Community Acts”.[23]Article 9 of the Directive provides that Member States shall postpone the adoption of a draft regulation for three months from the date of the communication of the draft to the Commission. This three month “stand-still” is extended where the Commission or another Member State delivers a detailed opinion to the effect that the measures envisaged may create obstacles to the free movement of goods within the internal market.

The background to the 2010 Regulations

[24]Three broad categories of devices which deliver a shock to the animal wearing the collar fall within the definition of “electronic collar” in the 2010 Regulations. They are usefully described in The Use of Shock-Collars and Their Impact on the Welfare of Dogs, an unpublished 2006 conference paper by Blackwell and Casey of the Department of Clinical Veterinary Science at the University of Bristol which reviewed the literature.[25]The first category consists of remote training collars operated manually by the trainer via a remote-controlled transmitter. The second category consists of devices that operate automatically in response to a dog barking. The third category consists of devices that are activated at a boundary line marked by a buried wire which interacts with the collar to keep the animal within a defined area. In paragraph 10 of his statement for the first judicial review Lord McNair stated that an electronic collar produces “3,000 times less electrical energy than allowed by standards for electrical fences”.[26]I have referred to the Kennel Club’s campaign for a ban on electronic collars and similar devices. On 6 December 2006 the then Welsh Minister for the Environment and Countryside and Professor Glossop met representatives of the Kennel Club to discuss the matter. In January 2007 the Minister and Professor Glossop met representatives of manufacturers of electronic training devices to discuss the matter from the industry’s perspective.[27]I have also referred to DEFRA’s project. Angela Critchley’s statement refers to an online article dated 31 August 2007 by one of a group of petitioners wanting a ban on electronic collars which set out the response the group received from the then Prime Minister. This inter alia stated:
“…There are no plans at present to ban the sale and use of any electronic training aids for animals, including the “electric shock collars”, and “while we are aware of a number of scientific studies on electric shock collars, DEFRA considers that to date those studies published in this area are not sufficiently robust and that the evidence base needs to be built on before consideration can be given to either banning or regulating their use…”
The article contains a link to the 10 Downing Street website. Angela Critchley states the response has been removed from the website, but that she was informed by Lord McNair that the quotation is accurate. Lord McNair’s evidence, however, does not refer to this article or to the then Prime Minister’s response.

and

[28]On 15 November 2007 the Welsh Ministers issued their first consultation paper. Paragraph 2 stated that the Welsh Assembly Government was “aware that there are strong views for and against the use of electronic training aids”. The paper stated that a number of animal welfare organisations believe they are cruel and unnecessary but others believe they can have an important role to play, particularly where an animal has proved unresponsive to other training techniques. It also stated that the second claimant believes that “using these collars on some dogs has saved some of their lives and has given them a far better quality of life”.[29]The consultation paper sought views as to whether any controls should be placed on electric shock or static pulse collars, whether their use, sale, distribution and possession should be banned, restricted, or whether they should continue to be allowed to be freely available. Those responding were asked to give their reasons for reaching a conclusion and for the evidence which led to it. That consultation included “scat-mats” as well as devices which used a collar.[30]The paper recognised (paragraph 23) that “if legislation is introduced to ban or regulate the sale or use of electric shock or static pulse collars there would be cost implications for those with commercial interests in these devices and possibly for some animal trainers”. It stated that, if proposals for a ban are to be made, it will be necessary for a Regulatory Impact Assessment to be prepared and that “in order to prepare a meaningful document it will be necessary for those who think that their business may be affected to provide as many details as possible of that [e]ffect”. It also stated that these details should include the financial consequences due to loss of business.[31]There were 173 responses to this paper. The responses came from a wide-range of individuals and organisations, including animal welfare groups, veterinary surgeons and their representatives, those involved in animal training, four chief constables and the four main political parties. The second claimant is listed as a respondent but the first claimant is not.[32]50% of those responding favoured a ban on all products including “stay-mats” and “scat-mats”. 39% disagreed and the remainder expressed no opinion. Those who supported a total ban on the use of electronic devices included a number of respected organisations in addition to the Kennel Club. The RSPCA, the National Search and Rescue Dog Association, the People’s Dispensary for Sick Animals, and the Pet Advisory Committee, (which consists of representatives of veterinary associations), the Local Government Association, the Pet Food Manufacturers Association and a number of animal welfare associations supported a total ban. The chair of the Welsh Association of Chief Police Officers (“ACPO”) stated that electric-shock collars had been banned by ACPO in 2000 and are not used by the police service.[33]Of those who did not favour a ban, the reasons given included:-(a) a ban did not make sense since the devices are available by mail order in England and other places in the EU;(b) the shock given cannot burn or scorch; and(c) the devices help to avoid injury to animals on roads or at the hands of farmers protecting their stock. I deal with the substantive reasons of those who favoured a ban, and the Chief Veterinary Officer’s assessment of the evidence and the responses in paragraphs [47]-[48] and [50] below.[34]As to the impact on manufacturers and distributors of regulation or a ban, the second claimant’s response to the consultation stated that the great detriment would be felt by 350,000 owners of animals who used or wished to use training collars. It gave an estimate of the number of electronic collars being used in Wales. This response was later summarised in the Explanatory Memorandum which was laid before the National Assembly on 24 February 2010 with the draft regulations: see [43]. The claimants’ letter before action, dated 24 March 2010 gave the same figures. It was only on 19 May, almost two months after the Regulations had come into force, that Ms Critchley, in her statement in support of these proceedings, provided further information. She stated (paragraph 7) that “it is difficult to gauge the effect because the ban has only been in effect for a short time” but “it has already had an impact and its continuing impact will clearly be severe”. The only evidence she tendered is the contents of an email from one of the first claimant’s customers. He stated:
“Comparing the six weeks prior to the ban with the six weeks past we have notice[d] a 40% drop in orders from Wales.”
In that period there were 6 online orders from Wales since the ban. On his figures that means that in the six weeks before the ban there were about 14 orders. He also stated that, on being informed of the ban, one customer cancelled, two changed to a spray option, two went ahead with the order, and a response was awaited from the sixth.[35]As to developments in Scotland, on 31 October 2008 a Member of the Scottish Parliament lodged a Parliamentary Question, asking the Scottish Executive when it planned to make regulations to ban the use of electric-shock collars. On 10 November 2008 the Scottish Minister for Rural Affairs and the Environment, Mr Richard Lochhead, stated that the responses to the 2007 consultation “showed that the arguments were finely balanced but inconclusive”. He also stated that “there is insufficient evidence at this stage to support a ban” and that a decision had been deferred until the results of the research by the Companion Animal Welfare Council and that by the Universities of Lincoln and Bristol commissioned by DEFRA was available.[36]In Wales, matters had by then proceeded differently. On 25 June 2007 the Minister, Elin Jones, made a statement to the National Assembly about the 2007 Welsh consultation and the Welsh Ministers’ approach. She stated:
“This has not been an easy subject to examine, as there is much emotion involved. There is a large degree of concern about these devices being improperly used, which is in contrast to responses from people who have used them and found that they have worked in stopping an animal from misbehaving. The industry that manufactures these devices obviously supports their continued use, while the veterinary profession has mixed views. In addition to all of this there are scientific publications that provide evidence against and for the acceptability of electric-shock collars.” … “After giving due considerations to the arguments, I propose to draft detailed regulations to introduce a ban on the use of electric-shock dog collars in Wales. Consideration will be given to whether their sale and possession should be included in the legislation. Appropriate provisions will be made to ensure that consideration is given to situations where the exceptional use of these devices could occur.”
[37]The representatives of the Welsh Conservatives, the Welsh Liberal Democrats and Plaid Cymru welcomed the Minister’s announcement. After they had expressed their support the Minister stated that the Welsh Ministers considered that “electric-shock dog collars are cruel and unnecessary”, and that they “are unnecessary because there are other means of training dogs”. She also stated that there may be situations where professionally supervised use of such devices may be justified. The DEFRA-funded research would not be completed until 2010 and a report was unlikely before 2011. While the Welsh Ministers would keep that research in view, she stated it was not their intention to wait for it. She made it clear that the products under consideration were those that give an animal a shock from which it cannot escape.[38]On 24 September 2008 the claimants launched their first judicial review. At that time draft regulations had not been published. The challenge was to the 25 June announcement. On 28 November 2008 permission was refused on the papers by David Holgate QC on the basis that the claim was premature. Draft regulations were published on 4 March 2009, but at the hearing of the renewed application for permission before HH Judge Jarman QC on 31 March permission was refused on the ground that the draft regulations were not in their final form so that the application was still premature.[39]The second consultation produced 51 responses from a range of people and organisations. They included veterinary surgeons, dog trainers and behaviourists, collar manufacturers and members of the public. 57% of the respondents favoured a complete ban on the devices and did not favour the defences, and 27% were against a complete ban. Sixteen percent did not give an opinion about the principle of a ban, but commented on the content of the draft regulations and the practicality of the proposed defences (summarised at [4]).[40]Professor Glossop (first statement paragraph 36) records that “local authorities, the police and professional bodies were sceptical that the defences would work”. This was particularly so in respect of the defence for the use of collars by or under the direction of a veterinary surgeon. The Royal College of Veterinary Surgeons and others considered that “veterinary surgeons may not have the necessary professional knowledge and expertise in the training of dogs to take such a decision”. Many respondents also believed the defence would not promote animal welfare. As to the defence proposed for using a device activated by proximity to a physical barrier for a short period, in the light of the consultation, it was concluded that use of such systems even for a short period was not consistent with the promotion of animal welfare.[41]The third consultation by the Welsh Ministers sought views about revisions to the draft regulations made in the light of the responses to the second consultation. The revised regulations were published on 21 December 2009. The consultation period ran until 8 February 2010. There were 18 responses to this consultation.[42]After the end of this consultation period, the Welsh Ministers decided only to prohibit the use of electronic collars and not devices such as “scat mats”. Draft regulations to this effect were laid before the National Assembly on 24 February 2010 together with the Explanatory Memorandum prepared by OCVO to which I have referred: [5]. The Minister stated that the Explanatory Memorandum gave a fair and reasonable view of the expected impact of the Regulations. Paragraph 4 of Part 1 was headed:
“Purpose & intended effect of the legislation”
. The material part stated:
“The purpose of the legislation is to prevent the use of electronic training/impulse collars on dogs and cats. The Animal Welfare Act 2006, implemented in Wales on 27 March 2007 introduced new concepts for the raising of standards of animals in Wales. Specifically Section 4 of that Act provides for an offence to have been committed if unnecessary suffering is caused, this include both mental and physical suffering. Commonly known as electronic shock collars or static pulse training devices, they are considered to conflict with positive reward-based training in that the “action” of triggering an electrical stimulus is not necessarily directly associated with the behaviour chain sought. Further, that the science to date suggests that the use of these collars adversely affects the behaviour of animals and in untrained hands could cause pain or distress. The legislation bans the use of these devices in Wales but not their sale or possession. By restricting the legislation to the act of using the devices, there are no trade issues and the issue of possession by a person transiting Wales does not arise…”
[43]The Memorandum also summarised the three consultations and their outcomes. Part 2 contains the Regulatory Impact Assessment. It stated there were four options:
“do nothing, provide appropriate non-statutory guidance, write a code of practice, and develop legislation”
. As to the first option, it stated: “Do Nothing Doing nothing is not an option. Section 9 of the Animal Welfare Act places a statutory duty on the person responsible for an animal to ensure its welfare. Allowing the continued use of electronic devices would be inconsistent with this duty given the evidence of harm being presented by both the animal welfare organisations, registration societies and indeed the industry.[44]Paragraph 7 dealt with costs and benefits. It stated:
“Immediate impact of change will fall on those persons who use these devices routinely. The consultation responses confirmed that in many cases these devices were only used briefly and saved them (the owners) time and effort. Whilst that is one view, it does not take account of the welfare of the animal. Annex 4 contains a list of the main areas of research which sustainably gives a view that there are problems with these devices, their usage and the resulting consequences on the animals. … The benefits of providing legislation include the certainty of being able to comply with the Animal Welfare Act 2006 and a certainty on the enforcement process. In Wales it is estimated 25% of households have a dog and 20% have cats. The 2001 census indicated that there were 1,209,000 households in Wales. Based on these figures, approximately 300,000 and 242,000 cats reside in Wales. The Electronic Collar Manufacturers Association (ECMA) estimate in 2006 that there were about 6,000 collars being used in Wales but they only have records for 414 collars sold in Wales. Other consultees (Kennel Club) confirm that there are unlikely to be great numbers of collars in use in Wales. However, ECMA, in the response to the third consultation say that they have evidence to suggest that the figures of collars used might be substantially higher and that “in 2010 there are likely to be in the region of half a million collars in use in the United Kingdom. It is not known how evenly spread the distribution is throughout the UK but a proportion of those collars are going to be in Wales, perhaps 50,000”
. These are subjective figures and the ECMA evidence is unsubstantiated by any sales, or other figures, although we do accept that the figures quoted in 2006 are now out of date. In terms of the number of jobs that may be put at risk, “ECMA estimate that the electronic training aid industry has a turnover of £2-3 million a year in the United Kingdom. It is not possible to say with certainty what proportion of this would apply to Wales. We are not aware of jobs in Wales that would be lost as a result of a ban”. Comments have been received in the third consultation from ECMA to suggest that the vested interests in training and behaviour are behind the bid for a ban on the use of these devices, but the training costs of a behaviourist are roughly [equivalent] to the cost of an electronic device. Devices can cost anywhere between £100-200, or even more depending on the product. Animal behaviourist costs are of the same order… …A number of people have commented that these collars have given them peace of mind. While that is noted the Animal Welfare Act only allows for the welfare of the animal to be considered. Therefore the question of benefit is whether the animal benefits. Modern thinking about positive training as encouraged by the veterinary professions, third section organisation, scientific research and in general animal behaviourists is sufficiently strong to recognise that these devices may cause problems for dogs and cats which cannot be ignored.”[45]As required by Directive 98/34/EC both sets of draft regulations were notified to the European Commission. The first draft regulations were notified on 5 March 2009. No observations, objections or representations were received from the Commission or other Member States. The second draft regulations were notified on 5 November 2009. On 6 November the Commission asked whether professional trainers would be exempt from the prohibition, what alternatives to electric collars and invisible fence systems had been considered as sufficient to train dangerous dogs, and whether devices which are not capable of causing shocks are excluded from the scope of the draft regulations. The United Kingdom, on behalf of the Welsh Ministers, responded to these questions on 26 January 2010. The answer stated that the reason for the ban was that the Welsh Ministers considered that the prohibition promoted the welfare of animals and that positive training, that is reward-based techniques, could meet both the welfare needs of an animal and the training requirements of its handler.[46]On 28 January the Commission asked for further details of the alternatives to electronic devices. The UK authorities responded on behalf of the Welsh Ministers on 3 February 2010, stating: “…There are a plethora of positive training methods available to suit all breeds and situations. Aggressive animals can be treated as part of behaviour modification programmes, which are based on an understanding of the motivation for each dog’s behaviour, and the use of humane, reward-based training methods. These can include lure-reward training, compulsion praise training and marker training.” (See Professor Glossop’s first statement, paragraph 48) Professor Glossop stated that no further comments were received from the Commission and three month stand-still period expired on 8 February 2010.[47]As to the reasons for, and the evidence in support of, a ban, the evidence is contained in Professor Glossop’s statements and the responses to the consultation. The material parts of Professor Glossop’s first statement are:
“26. Scientific research provides mixed evidence about the consequences of electronic collars for animals. It was my view that the balance of this scientific evidence indicated that electronic collars had the potential to have adverse consequences for animals, e.g. that they could cause harm to the animal, they could cause fear and anxiety for the animal which could in turn elicit aggression. The evidence also suggests there is a risk of “misassociation”
. This means that the dog associates the electric shock with a non-intended subject or behaviour. For example, a dog might associate the shock with the proximity of a human or another dog rather than with crossing the invisible boundary. 27. In my professional opinion, reward-based training and control using positive reinforcement is a better way of promoting the welfare of animals than the use of electronic devices that depend upon the delivery of an electric shock to an animal. There are a number of reasons for this conclusion. 28. First, positive reward-based training techniques have no negative physical impact on an animal, whereas the use of electronic collars inevitably involves the administration of an electric shock to animals. 29. Secondly, there are welfare implications resulting from the incorrect use of the electronic and the potential for abuse. Anxious and fearful responses can occur if the punishment is poorly synchronised with the action of the animal; and owners may not be objective in the use of punishment techniques. The potential for misuse exists particularly in those who are not fully skilled or experienced in using the collar, for example members of the general public (see, for example, Stress Symptoms, Caused by the Use of Electric Training Collars on Dogs: Schalke et al 2005). 30. Thirdly, positive reinforcement techniques help to address the underlying causes of a dog’s unwanted behaviour, whereas electronic collars simply suppress that behaviour. The behaviours for which people wish to use shocks are those that annoy humans yet those behaviours themselves can be either signals or signs of underlying distress in the animal. While the behaviour of the animal may be suppressed by electronic collars, the underlying cause of the behaviour is not addressed as a consequence.”[48]The RSPCA’s response to the consultation stated that no device should be used where an animal can be subjected to a painful stimulus at the direct instigation of a humanor where the stimulus is delivered as a result of an animal’s action from which it cannot retreat. The RSPCA considered that such collars could cause an escalation in the underlying behaviour and lead to aggression. The National Search and Rescue Dog Association, the PDSA and the other organisations to which I have referred [32] also stated that reward-based training techniques are preferable and better promote the welfare of animals.[49]There is an issue in the evidence as to whether the use of these devices causes “pain”. In his statement for the first judicial review Lord McNair stated (see paragraph 11) that “the subjective effects of low to medium [electronic stimulus] do not warrant the term “pain” as defined by the “International Association for the Study of Pain” because there is no actual or potential tissue damage. It is, however, not in dispute that the electric shocks cause discomfort and are of a sufficient intensity to affect an animal’s behaviour.[50]I turn to the scientific research. Ms Critchley exhibited a study by Steiss, Schaffer, Ahmad and Voith, “Evaluation of Plasma Cortisol Levels and Behaviour in Dogs Wearing Bark-Control Collars” (2007) 106 Applied Animal Behaviour Science 96. Professor Glossop had concerns about the reliability of this study and stated that other papers, summarised in the paper by Blackwell and Casey to which I have referred [24], suggested that the use of such collars is problematic. Her concerns about the reliability of the paper by Steiss et al are (see first statement paragraph 51) that the paper relied on measurement of plasma cortisol but, in recent years, the reliability of such measures as an indicator of the level of stress has been questioned. She also referred to the wide variation in normal cortisol levels for individual animals at rest and also to the variation in individual response to stress. Thirdly, the study relied on a lack of “statistically significant” increase in cortisol levels produced by an average over a group of dogs but the individual response may be masked within the group average and does not take into account the wide variation in individual response of animals to stress. Fourthly, the study used “shelter” dogs who may not react in the same way as non-shelter animals. Finally, the study relied on foot placement without explaining its significance although foot placement is not a commonly used behavioural indicator.[51]Dr Chamberlain (paragraphs 65-71) disagreed with Professor Glossop’s concerns. He also stated (paragraphs 59 and 72) that a number of the papers relied on by Professor Glossop used high powered electric collars which are now outdated. He also (paragraphs 74-77) criticised the review of the literature by Blackwell and Casey. His criticisms of it are that it did not assess “the full text” of each paper reviewed, did not relate the devices used in the studies to modern electronic collars, or the intended use of such collars, and that it miscited one of the papers. He also observed that it was funded by the RSPCA.[52]Professor Glossop stated that the research commissioned by DEFRA was originally expected to be completed by the beginning of 2010, but the timetable had slipped and it is unlikely to be completed before the end of 2010 or the beginning of 2011. Her first statement was made in July 2010, four months after the 2010 Regulations came into force. In it she stated (paragraph 58) that she was not aware of any information or updates concerning that project indicating a need for her to revise her professional opinion as to the best means to promote the welfare of cats and dogs. She also stated that the matter would be kept under review as evidence, whether from DEFRA or other sources, becomes available. She concluded (in paragraph 58):
“The evidence, to date, satisfies the Welsh Ministers that the use of alternative, reward-based training better promotes the welfare of animals than the use of collars delivering electric shocks. That is the conclusion that other countries within the European Union and outside it have also reached. I should also add that the use of collars has been banned by the police since 2000 and the armed forces do not use them…”
[53]As to whether an alternative to a ban was possible, in Professor Glossop’s professional opinion (first statement, paragraph 61) the alternatives suggested by the claimants would not adequately promote the welfare of animals. The criminal offence in section 4 of the 2006 Act is focused on the prevention of harm to an animal rather than promoting its welfare. The question whether an electronic collar caused unnecessary suffering would be a question of fact after the event in individual prosecutions, whereas the Regulations seek to promote animal welfare in a general way and before any harm is caused rather than purely protecting individual animals who have been harmed. Professor Glossop did not consider that dealing with the matter in a code of practice would suffice because a failure to follow a code of practice is not an offence and, since electronic collars could still be used, guidance in a code of practice would not serve to promote the welfare of the animals concerned.[54]Professor Glossop also referred to the complaint by the claimants, in effect a complaint of inconsistency, that electric fencing for livestock or equines has not been banned. This, she stated (first statement, paragraph 61(4)), failed to recognise the significant distinction between the visible barrier used for livestock and equines and the invisible barriers used in conjunction with electronic collars for cats and dogs. In the former, as a result of the visibility of the fence, any electric shock will be associated with the fence. There is, however, no such association between a shock and a boundary that is invisible.[55]Mr Chamberlain (statement, paragraphs 48 and 57) considered that two other alternatives showed that a total ban was disproportionate. The first was a training system for pet owners. The second was a licensing system which would permit those licensed to use electronic collars. Professor Glossop’s second statement (paragraph 13) observed that neither of these alternatives had been put forward during the consultation process. She also disagreed that they were satisfactory alternatives. A training or licensing system might reduce the potential for incorrect use of electronic collars, but could not affect the two other reasons for the Welsh Ministers’ conclusion. The first of those reasons was that the use of electronic collars inevitably involves the administration of an electric shock to animals whereas positive reward-based training techniques do not. The second was that electronic collars do not address the underlying cause of the unwanted behaviour but simply suppress it.[56]Professor Glossop also stated that the Welsh Ministers considered that a system of training was neither practicable nor enforceable. She gave as an example (second statement paragraph 21) that it would be difficult to prohibit owners giving or selling electronic collars to others, and it would be problematic to ensure that subsequent owners or users receive the required training. She also pointed to the responses to the proposal in the second consultation that there be defences for the use of electronic collars in exceptional circumstances including where the collar was attached by or under the direction of a veterinary surgeon: see [39] – [40]. That proposal did not find favour on consultation, inter alia for reasons of practicability. Discussion

The EU and ECHR grounds:

[57]Miss Carss-Frisk submitted on behalf of the claimants that the ban is unjustified and a disproportionate restriction on the free movement of goods contrary to Article 34 of the EU Treaty, and an unjustified deprivation of their possessions and a breach of Article 1 of the First Protocol to the European Convention on Human Rights. In respect of the latter she submitted that the claimants’ business interests and their ability to sell electric collars in Wales is a right in the nature of an asset which can be capitalised and thus an economic interest in the nature of goodwill which falls within the scope of Article 1 of the First Protocol. As Miss Carss-Frisk accepted that even if this is so, given the width of justification under Article 1 of the First Protocol, the claimants’ case based on Article 34 is stronger, I first consider that provision.[58]The starting point is to ask whether Article 34 is engaged. In the light of the decision of the Grand Chamber of the ECJ in Case C-110/05 Commission v Italy [2009] ECR 519 the answer is “yes” although the evidence of such interference in this case is exiguous compared with, for example, that in R (Countryside Alliance) v Attorney-General: see [2008] 1AC 219 at [25]. In Commission v Italy the ECJ considered Italian legislation which, on grounds of public safety, prohibited motorcycles from towing trailers designed for them. It held that the prohibition on towing such trailers was a measure having an equivalent effect to quantitative measures and required justification. It stated ([35]) that “obstacles to the free movement of goods which are the consequence of applying, to goods coming from other Member States where they are lawfully manufactured and marketed, rules that lay down requirements to be met by such constitute measures of equivalent effect to quantitative restrictions even if those rules apply to all products alike” and ([37]) that any measure “which hinders access of products originating in other Member States to the market of a Member State is also covered by that concept”. The ECJ also stated ([57]) that “the prohibition on the use of a product in the territory of a Member State has a considerable influence on the behaviour of consumers, which, in turn affects the access of that product to the market of that Member State”, and ([57]) consumers, knowing that they are not permitted to use a product “have practically no interest in buying” that product.[59]The ECJ had earlier stated ([34]) that what is now Article 34 of the Treaty “reflects the obligation to respect the principles of non-discrimination and of mutual recognition of products lawfully manufactured and marketed in other Member States, as well as the principle of ensuring free access of Community products to national markets”. It did not, as Lord Brown of Eaton-under-Heywood had hoped it would (see R (Countryside Alliance) v Attorney-General [2008] 1AC 719 at [149]) adopt “a remoteness test or de minimis requirement” for the application of Article 34 to non-discriminatory measures. Commission v Italy was followed in Case C-142/05 Åklagaren v Mickelsson and Roos [2009] ECR I-04 273 at [24] and [26].[60]So the issue is whether the ban is justified. This involves two or possibly three questions. The first is to identify the object of the measure, whether it is legitimate, and whether the measure is suitable to achieve a legitimate aim (the “suitability” test). The second (the “necessity” test) is whether the measure is necessary to achieve that aim, which involves consideration of whether there are less restrictive means of achieving it and, even if there are no less restrictive means, whether the measure has an excessive effect on the claimants’ interests. The “necessity” test can thus be seen as having two distinct elements although it is not always so analysed. What is clear is that it is for the national authorities to demonstrate that a measure that amounts to a derogation from the free movement of goods is consistent with the principle of proportionality: see Case C-170/04 Rosengren v Riksaklagaren [2007] ECR I-4107 at [50].[61]I discern from the Minister’s June 2007 statement and the Explanatory Memorandum that the Welsh Ministers had three areas of concern and that the purpose of the ban was the promotion of animal welfare. This is supported by Professor Glossop’s evidence. The areas of concern were that the collars administered an electric shock which causes discomfort and is of a sufficient intensity to affect an animal’s behaviour whether or not (see [49]) the discomfort satisfies the International Association for the Study of Pain’s definition of the term “pain”. Secondly the Explanatory Memorandum stated that in untrained hands the devices could cause pain and distress. Lord McNair’s statement in support of the earlier judicial review proceedings refers to Lindsay, Handbook of Applied Dog Behaviour and Training (vol. 3). At p.570 Lindsay stated that the devices are “potentially capable of producing significant psychological distress and harm when used improperly” but that when used “competently and selectively” they are “relatively safe and humane” (emphasis added).[62]Thirdly, it is stated that such devices conflict with positive reward-based training which provides positive reinforcement of conduct because the action which triggers the electronic stimulus is not directly associated with the behaviour sought. The Explanatory Memorandum and Professor Glossop’s first statement (paragraphs 26 and 30) state that electronic collars suppress the undesired behaviour without addressing the underlying causes and that because of this there is a risk of what Professor Glossop described (see [47]) as “mis-association” whereby the animal associates the electric shock with a non-intended subject or behaviour.[63]It is common ground between the parties that the promotion of animal welfare is a legitimate aim for the purposes of Article 34. It was, however, submitted on behalf of the claimants (see skeleton argument paragraph 41) that it is difficult to discern how the Regulations are alleged to fulfil that purpose. The submission is that it is difficult to discern whether the Regulations and the ban are a suitable means for protecting animal welfare. Miss Carss-Frisk QC said this was so because the main reason given by the Explanatory Memorandum and Professor Glossop for the ban is that such collars are aversive training techniques and that reward-based training techniques are better for animal welfare. She submitted that there is no evidence that, if electronic collars are banned, owners of animals and their trainers will use a reward-based method rather than another but non-banned aversive method. There were references to food deprivation and hitting which are not banned but in respect of which the protection for animals is by the offence under section 4 of the 2006 Act of causing unnecessary suffering to an animal. She also observed that other aversive techniques, for example scat-mats, were not banned.[64]The textual analysis undertaken by Miss Carss-Frisk of the differences between the Explanatory Memorandum, the Ministerial statement and the defendant’s Detailed Grounds of Defence does not help with regard to the purpose of the ban. That purpose was set out by the Minister and in the Explanatory Memorandum. The fact that other aversive techniques were not banned does not mean that the ban on electronic collars was not made to promote animal welfare or was not a suitable means. There was also a certain tension in Miss Carss-Frisk’s submission that the aim was not suitable in part because other aversive training techniques were not banned and her submission that it was disproportionate because other, less restrictive measures such as a licensing or training requirement had not been used[65]In the Countryside Alliance case it was argued that the rationale of the Hunting Act 2004 could not be, as the courts below had found, the prevention or reduction of unnecessary suffering to wild mammals overlaid by a moral viewpoint that causing suffering to animals for sport is unethical because if that was the purpose consistency would have required a more far-reaching measure. Lord Bingham rejected this argument. He described it as a traditional argument and stated that he did not “think that doubt can be thrown on the rationale of the Act, as expressed by the courts below, by showing that the underlying principle, if carried to its logical extremes, would have justified a much more far-reaching measure”: see [41]. He also stated (at [45]) that the legislative practice had been “whatever seemed at any given time to the current parliamentary majority to be the most pressing problem”. Although these paragraphs are in the section of Lord Bingham’s speech on Article 1 to the First Protocol, it is clear from paragraph [50] that he considered that this reasoning also applied to the analysis of what is now Article 34 of the Treaty and he did not draw a distinction between the “suitability” and the “necessity” part of the justification test in this regard. His observations are pertinent in the present context. Baroness Hale, in the context of Article 1 to the First Protocol, also rejected the “selectivity” argument, stating at [129] that protecting animals from avoidable compromise to their welfare fell well within the general interest and that the means chosen struck a fair balance. She stated “the fact that the same principles might have justified a wider ban does not mean that a narrow ban cannot be justified”. As to EC law, she agreed (see [131]) with Lord Bingham, and stated that the suggestion that the EC claimants in that case might succeed while the Human Rights claimants did not would be illogical and unjust.[66]As to the second limb of the justification test (necessity and no excessive effect) it is for the Member State (here the Welsh Ministers as the relevant national authority within the Member State) to demonstrate that its rules are appropriate and necessary to attain the legitimate object being pursued. But in Commission v Italy it was stated ([66]) “that burden of proof cannot be so extensive as to require the Member State [here the Welsh Ministers] to prove, positively, that no other conceivable measure could enable that objective to be attained under the same conditions”.[67]Miss Carss-Frisk criticised the defendant’s approach on three grounds. First, because it did not recognise the limit on the power to make regulations, which extends only to regulations in respect of animals “for which a person is responsible”. Secondly, the stated benefit of the legislation as including the certainty of being able to comply with the 2006 Act and certainty as to the enforcement process, assumed that certainty can only be provided by a ban. Thirdly, the Explanatory Memorandum considered that, because the legislation was not going to ban the sale or possession of electronic devices, “there are no trade issues”.[68]Taking the last of these first, Miss Carss-Frisk submitted that this indicates that Ministers did not undertake any detailed engagement with the considerations which should have been part of a proportionality analysis under European law. However, although she submitted that this meant that views of the defendant should be given less weight because it has not explicitly addressed the balancing exercise required, she accepted that the court is concerned with whether the ban is in fact justified, that is whether it is in fact unnecessary and thus a violation of Article 34. Although there was no explicit Article 34 proportionality analysis, it is clear from the Minister’s June 2007 statement and the February 2010 Explanatory Memorandum that those considering the policy did balance the perceived advantages of collars as put by those supporting them with their perceived disadvantages because of the evidence of harm and the impact on the welfare of the relevant animals.[69]As to the submission that it has not been shown that the ban will lead to an increased use of positive techniques, it will prevent the use of the aversive techniques the Welsh Ministers identified as the most harmful and problematic. It is of significance that during the course of the consultation, proposals to ban scat-mats were dropped in the light of the responses. It is also of significance that there was evidence that the use of electronic collars was capable of being misused or abused so as to cause pain and that one of the works relied on by the claimants (see [61]) accepted this. Dr Chamberlain considered that a training system or a licensing system would address this problem but these alternatives had not been suggested by the claimants during the consultation process. In any event they do not address two of the three reasons for the Welsh Ministers’ conclusion that the electronic collars should be banned. Those are:(a) that such collars inevitably involve the administration of an electric shock whereas positive reward-based training techniques do not, and(b) electronic collars do not address the underlying cause of the unwanted behaviour but simply suppress it. The reasons which led the Welsh Ministers to conclude that the defences they proposed in their second consultation (see [4], [40]) were not practical are also relevant in this context.[70]In any event, in the Countryside Alliance case Lord Hope stated ([78]) it was not necessary for those promoting the legislation to engage in a close and careful examination of the factual basis for concluding that hunting for foxes with hounds was less humane than other methods of killing them. His Lordship stated that it was open to the legislators to focus on the nature of the activities without comparing them with others, bearing in mind that they were being engaged in for sport and recreation. It was open to the legislators to form their own judgment as to whether they caused a sufficient degree of suffering in that context for legislative action to be taken.[71]Lord Hope’s analysis was in the context of Article 1 of the First Protocol. But his Lordship also stated [79], of Article 34 and European Community law, that it was within Parliament’s margin of discretion to address widespread concerns that had been raised about the hunting of foxes and to do so “within the limits imposed by the Treaty” by legislation.[72]Lord Hope also stated that, if it was open for the legislators to conclude that the activities that were to be prohibited were cruel if engaged in for sport, how could it be doubted that they were entitled to conclude that prohibition of those activities was necessary in order to prevent them from being carried on by those who wished to do so: [86]. That, he stated, was sufficient to meet what he referred to as the “condition of justification”.[73]In Commission v Italy the ECJ stated ([61]) in the absence of fully harmonising provisions at Community level, “it is for the Member States to decide upon the level at which they wish to ensure road safety in their territory whilst taking account of the requirements of free movement of goods within the European Community” and ([65]) in the field of road safety a Member State may determine the degree of protection which it wishes to apply in regard to such protection and the way in which that degree of protection is to be achieved and that that degree of protection may vary from one member state to the other and ([67]) that Member States cannot be denied the possibility of attaining an objective such as road safety “by the introduction of general and simple rules which will be easily understood and applied…and easily managed and supervised by the competent authorities”. The position is similar in relation to the decision of the Welsh Ministers as to the level at which they wish to ensure animal welfare. While the question whether there is a restriction within Article 34 is not subject to a de minimis principle, the extent of a restriction has a part to play in the assessment of proportionality.[74]In assessing whether the ban has been justified it is relevant that it was imposed after three rounds of consultation by Wales’s democratically elected representatives. The speeches in R (Countryside Alliance) v Attorney-General are of assistance. Lord Bingham stated that the existence of legislation by democratically elected representatives does not conclude the issue. However, it means that the issue is one in which “respect should be shown” to what the legislators decided. This, adapting his words to the facts of this case, is because “the democratic process is liable to be subverted if, on a question of moral and political judgment, opponents of [here the Regulations] achieve through the courts what they could not achieve in the [National Assembly for Wales]”.[75]Given the fact that the prohibition in this case was a measure of social policy aimed at animal welfare and not aimed at intra-Community trade, the situation is broadly similar to that in the Countryside Alliance case. Any impediment on trade between Member States was a minor and unintended consequence which (see Countryside Alliance, per Lord Bingham at [50], Lord Hope at [87] and Baroness Hale at [131]) bears more hardly on those within Wales than in other Member States. In these circumstances, I have concluded that the prohibition satisfies the requirement of proportionality under Community law and that the Regulations do not contravene that law.[76]While it is not conclusive that the notification process and the queries raised by the European Commission did not lead the Commission or another Member State to assert that the 2010 Regulations violate any provision of EU law, the fact that it has not done so is consistent with the reasoning that the ban of these collars pursues a legitimate aim and is necessary and proportionate. So to does the fact that such training collars are not permitted to be used in three other Member States, Finland, Denmark and Germany and that the Commission has not objected to the bans in those states or taken any enforcement action.[77]I also conclude that, assuming the ability to sell electronic collars is an economic interest in the nature of goodwill constituting a possession for the purposes of Article 1 of the First Protocol, that the prohibition on their use, in order to promote animal welfare is a justifiable interference because of the views taken by the Welsh Ministers as to what is appropriate in the interests of animal welfare. Miss Carss-Frisk in effect conceded that if the Welsh Ministers showed that the ban satisfied the European Community proportionality test, it would also satisfy the test under Article 1 of the First Protocol. Irrationality, Wednesbury unreasonableness and perversity:[78]As to the challenge to the Regulations in purely domestic public law terms, paragraph 39(2) of the claim form contends that the 2010 Regulations are “irrational, Wednesbury unreasonable and perverse.” Paragraphs 40-47 rely on the cumulative impact of six features. First, that section 4 of the 2006 Act makes it a criminal offence to cause unnecessary suffering to a protected animal or an animal for which one is responsible so that if a dog or cat is actually harmed by the use of an electronic collar the person responsible could be prosecuted. Secondly, the procedural powers, including the power to enter premises in sections 18-21 of the 2006 Act, which exist to ensure that any criminal harm to an animal does not go undetected and unpunished, and the power to provide in the relevant codes of practice that electronic collars are harmful are relied on to show that a ban is not necessary. Thirdly, it was submitted that it is perverse that electric fencing for livestock and equines which deliver a much more potent electric shock than an electronic collar are not banned. Fourthly, reliance was placed on the severe effect on the claimants’ trade, the failure of the Welsh Ministers to await the outcome of the DEFRA-funded research project, and their decision to proceed when the evidence in support of the proposition that electronic collars are harmful is unproven.[79]The factors to which I have referred in the context of Article 34 are also relevant in this context. In the light of the fact that the collars rely on the administration of an electric shock, that there are other methods of training or controlling animals which do not involve any negative physical impact on the animal and do, on the evidence before the Welsh Ministers, help address the underlying causes of the unwanted behaviour which the use of electronic collars suppress but do not address and the divided opinions of the scientists mean that the decision to ban the collars cannot be seen as irrational. The decision was supported by the majority of those responding to the consultation exercise including a number of respected animal welfare organisations and animal behaviourists. The Welsh Ministers only made their decision after receiving advice from the Chief Veterinary Officer for Wales and the decision has been approved by the democratically accountable and elected National Assembly for Wales.[80]The fact that those who cause unnecessary suffering to an animal may be prosecuted does not render the exercise of the rule-making powers to promote the welfare of animals irrational. This is because the criminal offences under section 4 and the other provisions address the position after the event. The rule-making powers enable regulations to be made which seek to promote animal welfare in a general way before any harm is caused to them rather than punishing those who have harmed individual animals. As to the argument that there should not be a distinction between barriers for livestock and equines and those for cats and dogs, it is not irrational to make a distinction between the effect of a visible barrier and the invisible barriers used in connection with electronic collars for cats and dogs: see the summary of Professor Glossop’s evidence at [54]. Moreover, the fact that the Regulations may have gone further and made the use of other devices, whether visible electronic barriers for livestock or scat-mats for cats and dogs, criminal does not demonstrate that the Regulations which were enacted are irrational.[81]The Welsh Ministers considered the commercial impact on those wishing to sell electronic collars (see the Regulatory Impact Assessment in the Explanatory Memorandum, the material parts of which are set out at [43] above), and decided that the importance of promoting animal welfare outweighed those considerations. While other relevant governmental authorities within the United Kingdom have taken a different decision, in Wales it is the Welsh Ministers and the National Assembly for Wales which is responsible. Given the outcome of the consultation in Wales and the advice of the Office of the Chief Veterinary Officer for Wales, it was not irrational or Wednesbury unreasonable to decide not to await the outcome of steps taken by the devolved administration in Scotland or the relevant authority for England before exercising the rule-making power under the 2006 Act. The new domestic law grounds:[82]Miss Carss-Frisk submitted that the regulations are ultra vires because Regulations 2(1)(a) and (b) are not restricted in their application to animals “for which a person is responsible” as required by section 12. Secondly, she submitted that Regulation 2(1)(c) has perverse consequences because it prohibits a person to “be responsible for a cat or a dog to which an electronic collar is attached” even if the purpose for taking responsibility is to care for the welfare of the animal or indeed to remove the electronic collar.[83]It was argued by Miss Carss-Frisk that the scheme of the Act distinguishes between various sub-groups of animals for the purposes of the different statutory controls, that section 12 authorises the making of regulations “for the purpose of promoting the welfare of animals for which a person is responsible” but the prohibition from attaching a collar or causing a collar to be attached is entirely irrespective of whether the person who attaches the collar or causes it to be attached is responsible for the cat or dog or whether there is anyone else who is in fact responsible for the cat or dog. Miss Carss-Frisk submitted that there is no ambiguity in the terms of the three prohibitions in regulation 2 which would enable the Regulations to be construed as only applying to animals for which a person is responsible. In short, the argument is that since the Regulations purport on their true and clear construction to apply, for example to stray dogs and to feral cats, they are ultra vires.[84]I reject Miss Carss-Frisk’s submissions. First, the prime guide to the meaning of delegated legislation is the intention of the legislature as indicated in the enabling Act: see Bennion on Statutory Interpretation 5th ed. 262-3, and Halsbury’s Laws of England vol. 44(1) paragraph 1522. Bennion refers to this as “the rule of primary intention” and Halsbury’s Laws states:
“The overriding principle in the interpretation of legislation made under powers conferred by statute is that it should be construed in the light of the enabling Act generally, and, in particular, so as to be consistent with its substantive provisions…”
[85]Bennion, in his comment to section 59 of his code, also states “the court will assume that the delegate intended to conform to the rule of primary intention, and will construe ambiguous provisions in the instrument accordingly”. Notwithstanding Bennion’s reference to “ambiguous provisions”, the breadth of the principle is illustrated by R v Croydon Crown Court, ex p Lewis (1994) 159 JP 886 in which the Divisional Court applied the principle to regulation 6(6) of the Prosecution of Offences (Custody Time-Limits) Regulations 1987. That provided: “The Crown Court on being notified that an accused who is in custody pending trial there has the benefit of a custody time limit…and that the time limit is about to expire, shall grant him bail in accordance with the Bail Act 1976 as from the expiry of the time limit, subject to a duty to appear before the Crown Court for trial”.[86]It was argued that the words “subject to a duty to appear before the Crown Court for trial” made it clear that bail granted at the end of the custody time limit was to extend to the date of the trial. Balcombe LJ rejected that argument. He did so because, if that was the true construction of regulation 6(6), it was clearly beyond the powers granted to the Secretary of State by the enabling provision in section 22(1) of the Prosecution of Offences Act 1985. Referring to the case of Raymond v Honey [1983] 1AC 1, Balcombe LJ stated “there is a well-established principle…that any regulations are to be construed in such a manner as that they shall not be ultra vires”. As it was possible to give regulation 6(6) a meaning which did not make it ultra vires the enabling Act because it did not expressly state that bail had to continue until the date of trial, he would so construe it. He did not refer to a requirement of ambiguity and the provision he was considering was no more ambiguous than regulations 2(1)(a)

(b) of the 2010 Regulations.

[87]The preamble to the 2010 Regulations states that they are made in exercise of the powers conferred by sub-sections (1), (3)(a) and (5) of section 12 and section 62 of the 2006 Act. I reject Mr Lewis’s submission that the reference to section 62 assists. It is the interpretation section and contains a number of definitions. The reference in the preamble may be to section 62(1) which makes the Welsh Ministers the appropriate national authority and gives them jurisdiction to make rules pursuant to section 12 rather than to section 62(4) which provides that references to “responsibility” in the 2006 Act are to be read in accordance with section 3. But the reference to the specified sub-sections of section 12 is a reference to provisions concerned only with animals “for which a person is responsible or the progeny of such animals”. The reference to progeny in the enabling power may itself possibly take the power beyond animals for which a person is responsible. But whether or not it does, in these circumstances the rule of primary intention and the overriding principle point to the construction of regulation 2(1)(a) and (b) as providing that a person commits an offence if that person attaches an electronic collar or causes such a collar to be attached to a cat or a dog for which a person is responsible or to the progeny of such a cat or dog. By section 3 responsibility includes ownership of the animal (section 3(3)), permanent or temporary responsibility (section 3(1)) and “being in charge of” an animal (Section 3(2)).[88]The counter-indication relied on by Miss Carss-Frisk is that regulation 2(1)(c) expressly prohibits a person “to be responsible for a cat or dog”. She submitted this shows that the prohibitions in sub-paragraphs (a) and (b) of Regulation 2(1) are not so confined. I do not consider this to be so. The three offences are; to attach an electronic collar, to cause an electronic collar to be attached, and to be responsible for a cat or dog to which such a collar is attached. There is no such need for an express reference to “responsibility” in sub-paragraphs (a) and (b) because the operative concepts in these provisions are ‘attachment’ and ‘causing to be attached’. They are concerned with the activity that is prohibited. They, for the reasons I have given, however, only apply to an animal for which a person is responsible. But Regulation 2(1)(c) is not concerned with an activity but a condition. It is difficult to see how the offence in (c) could be described without referring to the state of being responsible.[89]To construe the provisions of Regulations 2(1)(a) and (b) as extending to animals for which no person is responsible would also be inconsistent with section 271 of Bennion’s code; the principle against penalisation under a doubtful law. Section 271 states that, when considering “which of the opposing constructions of the enactment would give effect to the legislative purpose”, the court should “strive to avoid adopting a construction which penalises a person where the legislator’s intention to do so is doubtful, or penalises him or her in a way which was not made clear”. In Dickenson v Fletcher (1873) L.R. 9 CP 1 at 7 Brett J stated that those who contend that a penalty may be inflicted under a statute must show that its words distinctly so enact and that “they must fail if the words are merely equally capable of a construction that would, and one that would not, inflict the penalty”. Bennion does not discuss the interaction of his principle against penalisation under a doubtful law with the rule of primary intention which applies to delegated legislation but section 60 of his code states that “allowing for the difference in juridical nature and provenance, delegated legislation is to be construed in the same way as an act”, that is including the principle that avoids penalisation of conduct in cases where the intention to penalise is doubtful.[90]Construing regulation 2(1)(a) and (b) as applying to persons who act in the specified ways vis a vis a cat or a dog for which a person is responsible or the progeny of such a cat or dog as indicated in section 3 of the 2006 Act is consistent with both the rule of primary intention and the principle against penalisation under a doubtful law. In the present case the express reference to section 12 in the preamble to the 2010 Regulations shows that the construction limiting the ambit of regulation to a person responsible for the relevant animal is consistent with the intention of the legislature, here the National Assembly for Wales.[91]Regulation 2(1)(c) is not arguably ultra vires the enabling provision. It makes it an offence for a person to be responsible for a cat or a dog to which an electronic collar is attached. “Responsibility” includes (see section 3 of the 2006 Act) ownership, permanent or temporary responsibility, and “being in charge of” an animal. The argument that the provision is Wednesbury unreasonable or irrational because of its perverse consequences depends on an assumption that a person, not previously responsible, who “takes charge of” or takes hold of the animal in order to remove the electronic collar is caught by the provision. Given the purpose of the enabling provision and the 2010 Regulations that assumption involves an unreasonable and excessively literal construction. It is trite law that, in considering the unreasonableness of legislative acts such as regulations and by-laws made by public authorities, and in particular democratically accountable public authorities, a court should be slow to find that they are Wednesbury unreasonable or perverse.[92]The decision in Kruse v Johnson [1898] 2 QB 91 was given at a time when the scope of judicial review was far narrower than it is today. But Lord Russell of Killowen CJ’s statement about the general approach remains of assistance. It was, for example, applied in R v Dyfed CC, ex p. Mason [1995] Env. L.R. 83, 96. His Lordship stated that the court should jealously watch the exercise of powers by public authorities and guard against their unnecessary or unreasonable exercise to the public disadvantage. But he continued:
“[when] called upon to consider the by-laws of public representative bodies clothed with…ample authority…and exercising that authority accompanied by…checks and safeguards…the consideration of such by-laws ought to be approached from a different standpoint. They ought to be supported if possible. They ought to be, as has been said, “benevolently” interpreted and credit ought to be given to those who have to administer them that they will be reasonably administered…I think courts of justice ought to be slow to condemn as invalid any by-law so made under such conditions, on the ground of supposed unreasonableness.”
Courts have been particularly cautious where the legislative power is exercised by a Minister, and where the instrument has been laid before Parliament: see City of Edinburgh DC v Secretary of State for Scotland [1985] SLT 551 and R v Secretary of State, ex p. Nottinghamshire County Council [1986] AC 420; R v Secretary of State for Trade and Industry,ex p. Lohnro plc [1989] 1WLR 525, 536. Appropriateness of judicial review[93]My conclusions on the new grounds mean it is not necessary to consider Mr Lewis’s submission that it is inappropriate for this court to rule on these grounds in a judicial review brought by these claimants. He submitted that such matters should be dealt with within the context of criminal proceedings by way of a defence as in Boddington v British Transport Police [1999] 2 AC 143. This, he argued, is particularly so where the answer to the question will (as it is here) be fact-sensitive, and where (as is also the case here) the argument is first advanced at a very late stage. It is advanced in these proceedings without the presence of the animal welfare organisations who are most likely to be affected. I accept that it may well be appropriate for such welfare organisations as are involved in the prosecution of offences concerning animals, to be permitted to intervene, if they so wish, in a judicial review. But I consider that Mr Lewis’s general submission that judicial review is not appropriate in such cases faces formidable difficulties.[94]Boddington v British Transport Police was a case in which the House of Lords rejected the argument that a person is not entitled to raise the invalidity of regulations by way of a defence to a criminal charge where he has not sought to challenge the regulations in judicial review proceedings. It was not suggested in Boddington’s case that judicial review is not an appropriate remedy, only that it is not the exclusive remedy. Indeed, Lord Slynn and Lord Steyn contemplated referring the public law issues in a case in which the invalidity of regulations is alleged to the Divisional Court: see [1999] AC 143 at 164 and 176. Moreover, given the general effect of subordinate legislation, it will often be both appropriate and important for a substantive issue as to its validity to be determined at the outset rather than to wait for enforcement in the courts and then raising a collateral challenge. The generality of legislation and the fact that those affected by it and those responsible for enforcing it may rely on it from the time of its enactment has led courts on occasion to emphasise the need for promptness where a subordinate legislative instrument is challenged: see e.g. R v Secretary of State for Social Services, ex p. Association of Metropolitan Authorities [1986] 1 WLR 1, 15, albeit not in the context of the criminal law.[95]For these reasons, while permission is granted, this application is dismissed.

Cited in 2 later judgments