England and Wales High Court (Administrative Court) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		21/02/2018

Mr Justice Garnham:

Introduction

1. On 26 July 2017 the Department for Environment, Food and Rural Affairs ("DEFRA") published the "UK plan for tackling roadside nitrogen dioxide concentrations" and associated documents (hereafter "the 2017 Plan"). This was the third attempt by the UK Government to provide an Air Quality Plan ("AQP") that met its obligations in law.
2. The first AQP, produced in 2011, was quashed by order of the Supreme Court in 2015. The Government was made the subject of a mandatory order requiring the Secretary of State to prepare new air quality plans in accordance with a defined timetable (see R (on the Application of ClientEarth) v The Secretary of State for the Environment, Food and Rural Affairs [2015] UKSC 28, 4 All ER 724). The second AQP, produced in purported compliance with the order of the Supreme Court, was published on 17 December 2015.
3. In a judgment dated 2 November 2016 ([2016] EWHC 2740 (Admin), ("the November 2016 judgment"), I held that the 2015 plan was also deficient. I made a direction that DEFRA must publish a new AQP, which complied with the relevant EU Directive and domestic Regulations, by 31 July 2017. It was in purported compliance with that order that DEFRA published the 2017 Plan.
4. The Claimant in these proceedings is "ClientEarth", a registered charity, whose objects include promoting and encouraging the "enhancement, restoration, conservation and protection of the environment, including the protection of human health, for the public benefit". By these proceedings, the Claimant challenges the 2017 Plan on the ground that it too failed to meet DEFRA's legal obligation. ClientEarth was also the claimant in the two previous judicial review cases. The Defendants are the Secretaries of State for Food, Environment and Rural Affair, and for Transport, and the Welsh Ministers. The Secretary of State for Food, Environment and Rural Affair has taken the lead for the Defendants in this case (and I refer to him hereafter as "the Secretary of State").
5. Proper and timely compliance with the law in this field matters. It matters, first, because the Government is as much subject of the law as any citizen or any other body in the UK. Accordingly, it is obliged to comply with the Directive and the Regulations and with the orders of the court. Second, it matters because, as is common ground between the parties to this litigation, a failure to comply with these legal requirements exposes the citizens of the UK to a real and persistent risk of significant harm. The 2017 Plan says that "poor air quality is the largest environmental risk to public health in the UK. It is known to have more severe effects on vulnerable groups, for example the elderly, children and people already suffering from pre-existing health conditions such as respiratory and cardiovascular conditions". As I pointed out in the November 2016 judgment, DEFRA's own analysis has suggested that exposure to nitrogen dioxide (NO₂) has an effect on mortality "equivalent to 23,500 deaths" every year.

The Legislative Scheme

6. At paragraphs 6-15 of the November 2016 judgment, I set out and explained the legislative background relevant to the arguments in that case. That background remains relevant to this challenge but it is not necessary to repeat all that detail here. It suffices for me to note the following provisions, all of which are cited in the November 2016 judgment:
7. Article 13 of Directive 2008/50/EC ("the 2008 Directive") imposes limit values and alert thresholds for the protection of human health. It provides:

"1. Member States shall ensure that, throughout their zones and agglomerations, levels of sulphur dioxide, PM10, lead and carbon monoxide in ambient air do not exceed the limit values laid down in Annex XI.

In respect of nitrogen dioxide and benzene the limit values, specified in Annex XI may not be exceeded from the date specified therein."

8. Article 23 provides that:

"Where, in given zones or agglomerations, the levels of pollutants in ambient air exceeds any limit value…member states shall ensure that air quality plans are established for those zones and agglomerations in order to achieve the related limit value…specified in Annexes XI and XIV.

In the event of exceedances of those values for which the attainment deadlines have already expired the air quality plan shall set out appropriate measures, so that the exceedance period can be kept as short as possible."

9. Annex XI to the 2008 Directive imposes a limit value for nitrogen dioxide of an average of 200ug/m³ in any given hour (which is not to be exceeded more than 18 times in a calendar year) and an average of 40ug/m³ which applies to each calendar year.
10. Annex XV sets out information to be included in the local, regional or national air quality plans for improvement in ambient air quality. Amongst the information required is detail of those measures or projects adopted with the view to reducing pollution. The Plan must list and describe all the measures set out in the project, set out a timetable for implementation, provide an estimate of the improvement of air quality planned and the expected time required to obtain that objective.
11. The 2008 Directive was brought into domestic law in the UK by means of four sets of regulations, one for each of the home nations. Regulation 26 of the English Regulations (the Air Quality Standards Regulations 2010 (2010/1001)) requires the Secretary of State, when the levels of nitrogen dioxide (amongst other pollutants) exceeds any limit value, to draw up and implement an AQP so as to achieve that limit value.
12. Regulation 26 also specifies that the AQP must "include measures intended to ensure compliance with any relevant limit value within the shortest possible time…." and "must include the information listed in Schedule 8."
13. In addition to the provisions referred to in the November 2016 judgment, it is material to note the following five additional provisions.
14. First, paragraph 8 of Schedule 8 (which, as noted above, is referred to in Regulation 26 of the English Regulations) specifies, as part of the information which must be included in air quality plans, the following:

"Details of those measures or objectives adopted with a view to reducing pollution following 11 June 2008 - (a) listing and description of all the measures set out in the project; (b) the timetable for implementation; (c) estimate of the improvement of air quality planned and of the expected time required to attain these objectives"

15. Second, the Air Quality Standards (Wales) Regulations 2010 (2010/1433) impose, on the Welsh Ministers, duties in respect of Wales equivalent to those imposed on the Secretary of State in respect of England. In particular, regulations 13 and 20 of those Regulations provide:

"13.(1) …the Welsh ministers must ensure levels of …nitrogen dioxide…do not exceed the limit values set out in Schedule 1 in any zone…

(Schedule 1 imposes the same limit values as are imposed in England.)

20. Where the level of …nitrogen dioxide…in ambient air exceeds any of the limit values in Schedule 1 in any zone…the Welsh Ministers must draw up and implement an air quality plan to achieve the relevant limit value…in that zone."

16. Third, Articles 6-7 of the 2008 Directive makes provision for assessment criteria and sampling points in order to ensure consistent monitoring of ambient air quality across the EU.
17. Fourth, in December 2011, the European Commission published a Commission Implementing Decision laying down rules for the 2008 Directive as regards reporting of ambient air quality. Paragraph 1 of Article 13 of that decision provides that:

"Member States shall make available the information set out in Parts…K of Annex II to this Decision on air quality plans as required by Article 23 of Directive 2008/50/EC including (a) the mandatory elements of the air quality plan as listed pursuant to Article 23 of the Directive 2008/50/EC in Section A of Annex XV to Directive 2008/50/EC…"

18. Part K requires the provision of information as to matters including:

"(14) Planned implementation: start and end date

(15) Date when the measure is planned to take full effect

(16) Other key implementation dates

(17) Indicator for monitoring progress

(18) Reduction in annual emissions due to applied measure…"

19. Finally, s85 of the Environment Act 1995 provides, as material:

"(3) If it appears to the [the Secretary of State]

(a) that air quality standards or objectives are not being achieved, or are not likely within the relevant period to be achieved, within the area of a local authority…

the appropriate authority may give directions to the local authority requiring it to take such steps as may be specified in the directions…

(7) It is the duty of a local authority to comply with any direction given to it under or by virtue of this Part."

20. It is against that statutory framework that the new AQP was developed.

Developing the 2017 Air Quality Plan

21. Shortly after the November 2016 judgment, DEFRA set about the task of preparing a new AQP.
22. I was provided with a detailed and helpful statement from Mr Andrew Jackson, Deputy Director of a unit established by DEFRA and the Department for Transport and known as the Joint Air Quality Unit ("JAQU"). It is apparent from the statement that very considerable time and effort was devoted to the preparation of the plan by officials and ministers.
23. A long list of potential policy options to tackle nitrogen dioxide emissions was identified; strategy papers analysing the problem were produced; proposals were discussed with the Greater London Authority, other local authorities and the devolved administrations. Amongst the options was a wider deployment of clean air zones ("CAZs") than that contemplated by the 2015 plan.
24. In 2016 the Government had published a "Clean Air Zone Framework in England" which set out the principles for the operation of a CAZ. As was subsequently to be explained in paragraph 103 of the 2017 Plan, CAZs fall into two categories:

"a. Non-charging Clean Air Zones – These are defined geographic areas used as a focus for action to improve air quality. This action can take a range of forms including, but not limited to, those set out in Section 2 of the Framework but does not include the use of charge based access restrictions.

b. Charging Clean Air Zones – These are zones where, in addition to the above, vehicle owners are required to pay a charge to enter, or move within, a zone if they are driving a vehicle that does not meet the particular standard for their vehicle type in that zone. Clean Air Zone proposals are not required to include a charging zone, and local authorities may consider alternatives to charging such as access restrictions for certain types of vehicles."

25. As a result of the work described above it was determined, consistent with the analysis referred to in the November 2016 judgment, that "Charging Clean Air Zones" (or "Charging CAZs") were the preferred option for reducing roadside NO₂ emissions.
26. By the middle of April 2017, a draft AQP was nearing completion. DEFRA sought an extension of time from the Court for the production of the draft plan because of the approach of local, and then national, elections. I granted a modest extension to cover the local election, but refused a much longer one in respect of the general election ([2017] EWHC 1618 (Admin)). The date of the publication of the final report was maintained at 31 July 2017.
27. In accordance with the amended order of the Court, a draft air quality plan and supporting technical report were published on 5 May 2017. Those documents were then put out to consultation. In June 2017 ClientEarth sought an order that the Secretary of State should produce a supplement to the draft published in May 2017. I refused that application ([2017] EWHC 1966 (Admin)).
28. The consultation process ended on 15 June by which time some 743 substantive responses had been received. The responses included a substantive one from ClientEarth and a number of pro forma responses from members of the public who were encouraged to respond by ClientEarth. A summary of the responses to the consultation was published subsequently.
29. The period between the end of the consultation period and production of the final AQP was marked by high level meetings between officials and ministers at which decisions were made as to the final shape of the report. On 26 July 2017 the Government and the devolved administrations published 3 documents. First, the "UK plan for tackling roadside nitrogen dioxide concentrations: an overview" (also known as "the Overview Document"). Second, the "UK plan for tackling roadside nitrogen dioxide concentrations: Detailed plan" ("the Detailed Plan"). Third, the technical report.
30. On 27 July 2017 the Government published a Direction to 23 local authorities under s85(5) of the Environment Act 1995. This Direction, entitled "The Environment Act 1995 (feasibility study for nitrogen dioxide compliance) air quality direction 2017", required the 23 authorities to undertake a feasibility study to identify the option which will deliver compliance with legal limits for nitrogen dioxide in the area for which the authority is responsible, in the shortest possible time.

The 2017 Air Quality Plan

31. Identification of what are the critical elements of the new air quality plan is not contentious and I am able to summarise the position in relatively short compass.

Zones, Local Authorities and National Measures

32. Section 3 of the Detailed Plan explains that the UK is divided into 43 zones for air quality reporting. In all but two zones, the UK is achieving the statutory hourly mean limit value for NO_2. However, 37 zones exceeded the statutory annual mean limit value for NO₂ in 2015.
33. These zones are not co-terminal with local authority areas; many zones incorporate more than one local authority area. For the purposes of its operation, however, the 2017 Plan is directed to local authorities, who are to have a central role in bringing the plan into effect. Section 7.4.1 of the Detailed Plan outlines the requirement for local authority-led action plans in England. Paragraph 90 of the Detailed Plan provides that:

"Given the local nature of the problem, local action is needed to achieve improvements in air quality. As the UK improves air quality nationally, air quality hotspots are going to become even more localised and the importance of action at a local level will increase. Local knowledge is vital to finding air quality solutions that are suited to local areas and the communities and businesses affected. A leading role for local authorities is therefore essential."

34. It is acknowledged that locally-led solutions will need to be implemented within a national framework designed to ensure that compliance will be achieved within the shortest possible time.
35. The 2017 Plan identifies a range of existing actions that were already being taken to tackle local NO₂ exceedances and reduce overall emissions. First, there is action taking place across the UK, including, for example, improvements to emissions testing for vehicles. Second, there is action being taken in England: for example, action by Highways England to improve air quality on the strategic road network in England, and action to update Government procurement policy to encourage the purchasing of cleaner vehicles by Government.

Non-Compliant Areas and Annex K

36. The 2017 Plan explains that on 31 July 2017 the Government published 37 individual zone plans for each non-complaint zone in the UK. I return to the contents of the local plans below.
37. The degree of non-compliance exhibited and forecast for different local areas varies widely. Annex K to the Detailed Plan sets out these different forecasts by reference to local authorities. Different approaches are adopted in the Plan depending on the degree of non-compliance forecast.
38. First, there are 23 local authorities representing areas with the greatest problem, i.e. those with exceedances projected beyond the next three to four years. Second, there are the five cities that were previously the focus of the 2015 AQP (Birmingham, Leeds, Nottingham, Derby and Southampton). Third, there are 45 local authorities which currently have air quality exceedances, but which are expected to achieve compliance with the NO₂ limit values by 2021.
39. Although different measures are planned for each of these groups, the plan explains that "the UK government has identified Clean Air Zones that include charging as the measure it is able to model nationally which will achieve statutory NO₂ limit values in … the shortest possible time". Accordingly, this measure is to be used as "the benchmark for assessing locally-led solutions".
40. For the first group, those areas with exceedances projecting beyond the next three to four years, local authorities are required to develop local plans in order "to achieve the statutory NO₂ limit values within the shortest possible time". Paragraph 111 of the Detailed Plan explains that if local authorities adopt a Charging CAZ, modelling suggests that they could achieve statutory NO₂ limit values in most cases by 2021. That allows for the time needed to design, commission and install CAZs and bring them into operation.
41. Given the potential impacts on individuals and businesses of CAZs and other measures, the Plan provides that if local authorities can identify measures other than Charging CAZs, which are at least as effective at reducing NO₂, then such measures are to be preferred. However, the local authority must demonstrate that these will deliver compliance as quickly as a Charging CAZ. The Government will only approve local authority plans if the local authority can show that its plan is likely to cause NO₂ levels in the area to reach legal compliance within the shortest time possible (and that it provides a route to compliance which reduces exposure as quickly as possible). By virtue of the July 2017 Direction, these local authorities are subject to legal duties to develop and implement such local plans.
42. The relevant 23 local areas are required to develop local plans and implement them "at pace" so that air quality limits are achieved within the shortest possible time. Specifically, they are required to set out initial plans by the end of March 2018, at the latest, and final plans by the end of December 2018 at the latest.
43. A somewhat different approach is taken to the second group, the five cities that were previously the focus of the 2015 AQP. That AQP anticipated that the five cities would be mandated to implement Charging CAZs which would achieve compliance by 2020. Consequently, paragraph 112 of the 2017 Plan makes clear that:

"The UK government continues to expect local authorities in the five cities named above to deliver their Clean Air Zones by the end of 2019, with a view to achieving statutory NO₂ limit values within the shortest possible time, which the latest assessment indicates will be in 2020."

44. The 2017 Plan provides that the five cities are working to the same timetable as they were under the 2015 AQP. A more detailed breakdown of the proposed timetable was set out in the draft Technical Report published in May 2017. The Secretary of State contends that JAQU and Defra have been engaging, and continue to engage, intensively with each of the five cities and have been closely supporting them in the development of their plans for achieving compliance. JAQU has provided feedback on the Outline Business Cases submitted by the five cities to date.
45. On 19 December 2017, new Directions were issued to each of the five cities under s85(5) of the Environment Act 1995 requiring the relevant local authority to prepare, as part of its feasibility study, a full business case for the area for which it is responsible, which was to be submitted to the Secretary of State as soon as possible, and by 15 September 2018 at the latest. JAQU has indicated to the cities that it is intended subsequently to use Ministerial Directions to direct each local authority to implement its local plan (full business case) once it has been approved by the Secretary of State.
46. The third group are the 45 local authorities which currently have air quality exceedances, but which are expected to achieve compliance with the NO₂ limit values by 2021. The 2017 Plan proceeds on the basis that these local authorities are not required to develop further local plans or undertake a feasibility study benchmarked against a Charging CAZ.
47. The situation as regards these local authority areas is not homogenous. Of these 45 local authorities, 12 are expected to achieve compliance in 2018, a further 10 are expected to achieve compliance in 2019, a further 13 are expected to achieve compliance in 2020, and the remaining 10 are expected to achieve compliance in 2021.
48. The Detailed Plan explains that the implementation of a CAZ is expected to take up to three years. Paragraph 116 provides that the government

"will only require local authorities to develop plans where evidence suggests measures could be put in place to bring forward achievement of statutory NO₂ limit values".

49. However, the Plan says that the government is conscious that some local authorities, namely these 45, are forecast to have air quality exceedances "which are close to, but below air quality limits in 2021" and therefore it

"will consider further steps to ensure that air quality in these areas improves and to ensure that forecast levels remain compliant. These steps could include preferential access to funding and government support to access and build on best practice."

50. The Technical Report also explains that:

"Those areas with the greatest problem, with exceedances projected beyond the next three to four years, will be required to develop local plans. Other areas will also be expected to take steps now to reduce emissions if there are measures they could take to bring forward the point where they meet legal limits and government will take steps to support them."

51. The Secretary of State asserts that, depending on the extent and source of the exceedances, different local authorities are adopting different policies and measures to address air quality issues. He says that JAQU has undertaken a review of the situation in these areas which the unit proposes to share with them to help them to focus their efforts. It is said that all 45 local authorities can also access support from DEFRA as part of the Local Air Quality Management (LAQM) framework, including a dedicated LAQM helpdesk.

Wales

52. The Welsh Ministers are responsible for those parts of the 2017 Plan which fall within their devolved competence and for which they have been designated the competent authority for the purposes of Directive 2008/50/EC. The Welsh AQP primarily consists of the Detailed Plan and the zone plans for the four Welsh air quality zones.
53. For reasons that will become apparent, I need to say no more about background to the claim against the Welsh Ministers.

The Competing Arguments

54. I had the benefit of detailed skeleton arguments from Nathalie Lieven QC and Ravi Mehta on behalf of the Claimant, and Kassie Smith QC and Julianne Morrison for the Secretary of State. I also heard careful and well-structured oral submissions from Ms Lieven and Ms Smith. I am grateful to all Counsel and to those who instruct them for the manner in which this case has been prepared and argued. I do not intend to do more here than summarise the parties' respective arguments; the skeletons provide a more detailed overview of their cases.
55. Ms Lieven advanced two principle grounds in support of her contention that the 2017 Plan is unlawful in respect of England.
56. In her skeleton argument she summarised her first argument by saying that "a substantial number of local authority areas in England are unaccounted-for." She went on to develop that argument in rather less bald terms. She says that in relation to 45 local authority areas in England, the AQP "includes no concrete, impact-assessed measures to ensure compliance in the 'shortest possible time', nor any requirement for responsible local authorities to "carry out feasibility studies or to identify such measures, despite identifying ongoing breaches of limit values".
57. Ms Lieven says that the adoption of a benchmark provided by Charging CAZs is misplaced in the case of these areas because it avoids the obligation to ensure compliance in the "shortest possible time". She says that the 45 Local Authorities will not have the same access to funding as the local authorities who are included in the Direction. She says that the Individual Zones Plans contain lists of measures designed to ensure compliance with legal limit values, but with "largely unquantified impacts". She says that no timeline is given for additional measures to be taken in the 45 local authority areas, no concrete measures are identified and no indication is given of the likely improvements from those steps. She says that in any event projected compliance is based on over-optimistic modelling.
58. Ms Lieven's second ground relates to provision made in the 2017 Plan for the five cities alongside London that were previously to be mandated to introduce CAZs (Birmingham, Leeds, Nottingham, Derby and Southampton). She contends that the Detailed Plan originally imposed no legal requirement for the timing or scope of their introduction of Charging CAZs. She says that "the 2017 Directions effectively concede part of this claim". Nonetheless she argues that these directions do not meet the requirements of EU law for a clear and legally enforceable timetable for implementation of the necessary measures.
59. In response, Ms Smith argues that the first complaint is "misconceived". The 45 local authority areas are not "unaccounted-for". In each area, action is being taken to address air quality issues. The individual zone plans for the areas covered by the 45 local authorities set out the measures that have been implemented to date, or are planned and being taken in each area to reduce NO₂ levels within a reporting Zone.
60. She says that the government has identified Charging CAZs as the measure that will achieve compliance with the NO₂ limit values in the shortest possible time and the benchmark against which any local authority plans will be assessed. Given the projected timeframe for compliance in each of these areas, the introduction of CAZs would not bring forward compliance. Consequently, she argues it would be disproportionate and inappropriate for these areas to be mandated to take steps towards introducing one. In particular, she contends, the preparation of feasibility studies and necessary local consultation is not expected to identify measures that could be worked up and introduced in time to bring forward compliance.
61. This does not mean, Ms Smith contends, that no further action will be taken in these areas. In particular, in those areas which are forecast to have air quality exceedances which are close to, but below air quality limits in 2021, as well as the matters set out in the individual zone plans, the Government "will consider further steps to ensure that air quality in these areas improves and to ensure that forecast levels remain compliant. These steps could include preferential access to funding and government support to access and build on best practice." She says that JAQU is already engaging with relevant authorities in order to identify what further steps can be taken to support them.
62. Ms Smith says that the national monitoring and modelling used for the purposes of the 2017 Plan has been undertaken in accordance with the criteria set out in and the requirements of the Air Quality Directive.
63. According to Ms Smith, ClientEarth's second complaint is also misconceived. The Joint Air Quality Unit, she says, is working intensively with the five cities to ensure that they deliver their CAZs to the timetable anticipated by the 2017 Plan (i.e. CAZs to be in place by the end of 2019, achieving compliance in 2020). She argues that ClientEarth is wrong to contend that the Plan can only be effective if the Secretary of State imposes mandatory timetabling requirements, addressing all stages of the process, on the five cities from the outset.
64. In any event, she says, that is not required by the Air Quality Directive. Moreover, she says, the 2017 Plan always envisaged mandating authorities to act to implement their measures in accordance with the timetable outlined in the 2015 AQP. Legally binding Ministerial Directions have now been issued to the five cities to submit their full business cases to the Secretary of State by 15 September 2018, and Directions will subsequently be issued requiring each of the five cities to implement its local plan, as set out in its full business case, once it has been approved by the Secretary of State.
65. Ms Smith disputes ClientEarth's contention that the decision to issue the December Directions concedes part of its claim. Instead, the issuing of the 2017 Directions demonstrates that the Secretary of State is continuing to work to ensure that the five cities achieve compliance as soon as possible. She says that the change in the means of applying obligations (a move from a Statutory Instrument to legally binding Directions) does not assist the Claimant's case. The use of Directions is predicated on the need for a tailored, timely and focused approach.

Discussion

66. Central to the argument as it was developed at the hearing was Table 1 of Annex K to the Detailed Plan, which provides a summary of proposed remedial measures. That table identifies local authorities in England "with roads with concentrations of NO₂ forecast above legal limits and assuming no additional measures". It is possible to identify from that table three categories of local authority.
67. The first consists of the Greater London Authority (the plans in respect of which are not challenged in this case) and the five cities of Birmingham, Derby, Leeds, Nottingham and Southampton (excluding a single stretch of road in the New Forest). The 2015 Plan assumed a Clean Air Zone was required in each of these areas. The second consists of 23 local authorities (including New Forest District Council but excluding Halton Borough Council where the opening of the Mersey Gateway Bridge was thought likely to solve the problem), which are to be "required to produce local action plans by March 2018". The third is the 45 local authorities which are "not required to conduct a feasibility study". Ms Lieven's first ground focuses on the third category and her second on the first category.

The 45

68. It is perfectly plain that the 45 local authorities are not "unaccounted for" as Ms Lieven's skeleton asserted. On the contrary, they are expressly identified in Table 1 and discussed in paragraph 116 of the Detailed Plan which I have set out above.
69. It is equally apparent, however, that the fact that these 45 local authority areas are expected to achieve compliance with the statutory NO₂ limit values by 2021 has led the Government to impose on them less onerous obligations than is the case for the 28, namely the five cities and the 23 other authorities (plus London) in respect of which compliance will not be achieved until after 2021. It is also plain that the reason for this distinction is the Government's assessment that these 45 will become compliant, without further measures being taken, within the period of three years which it would take to design, install and bring into operation a Charging CAZ.
70. Whilst no concession is made, no real point is taken on the assertion that it would take three years to introduce a Charging CAZ, nor on the assertion that Charging CAZs are the most effective means of addressing NO₂ exceedances. Nor can it be said that there is any error of approach in the government adopting Charging CAZs as the yardstick against which any alternative scheme is to be tested. In consequence, there is no challenge in this regard to the proposals in the 2017 plan in respect of the 23 authorities or to the plan to introduce CAZs in the five cities.
71. But where, in my judgment, the Government's plan is flawed, and seriously flawed, is in its application of the 3 year benchmark to the 45 local authority areas where compliance is anticipated within 3 years in any event. Plainly, it would be pointless to require these local authorities to embark on the expensive and time consuming enterprise of establishing a CAZ in an area where compliance will be achieved within the same period without a CAZ. But the Government cannot sensibly, or lawfully, substitute the application of its benchmark, however rational in respect of areas where a CAZ is the most efficacious solution, for the requirements of the Directive and the Regulations in areas where it is not.
72. The obligation imposed by Article 23 of the 2008 Directive is specific to each and every zone or agglomeration. The obligation to devise air quality plans applies "where, in given zones or agglomerations, the levels of pollutants in ambient air exceed any limit value" (emphasis added). When the obligation arises the Article requires Member States to ensure that AQPs are established "for those zones".
73. As I explained in the November 2016 judgment, the proper construction of Article 23 imposes a three-fold obligation on the Secretary of State; he must aim to achieve compliance by the soonest date possible; he must choose a route to that objective which reduces exposure as quickly as possible; and that he must take steps which mean meeting the value limits is not just possible, but likely. It follows that the Secretary of State must ensure that there is in place a plan for each zone which meets the three-fold obligation.
74. Because the obligation is zone-specific, the fact that each of the 45 local authority areas will achieve compliance in any event by 2021 is of no immediate significance. The Secretary of State must ensure that, in each of the 45 areas, steps are taken to achieve compliance as soon as possible, by the quickest route possible and by a means that makes that outcome likely. The CAZ benchmark cannot be treated as a means of watering down those obligations.
75. Nor is it an answer to this point to say, as Ms Smith does, that the current plan, with its careful application of the CAZ benchmark, is a "proportionate" response by the government to the issue raised by NO₂ emissions. Implicit in that submission is a suggestion that cost may play a part in determining the national AQP; that when viewed as a whole, the 2017 Plan is reasonable because it demands expenditure and action where there are exceedances that will persist, but demands less when the effluxion of time will bring zones into compliance without such costs. I reject that argument.
76. For the reasons I explained at paragraph 50 of the November 2016 judgment, the obligations imposed by the 2008 Directive are not qualified by reference to their cost:

"I reject any suggestion that the state can have any regard to cost in fixing the target date for compliance or in determining the route by which the compliance can be achieved where one route produces results quicker than another. In those respects the determining consideration has to be the efficacy of the measure in question and not their cost. That, it seems to me, flows inevitably from the requirements in the Article to keep the exceedance period as short as possible."

77. In consequence, the expression "proportionate" has a very particular meaning in the present context. I stand by the definition of that word offered in the November 2016 judgment: "the measures a Member State may adopt should indeed be "proportionate", but they must be proportionate in the sense of being no more than is required to meet the target". I note that DEFRA chose not to appeal the 2016 decision. Because the target in view is compliance with the 2008 Directive in all zones, the expense of doing so promptly in any one zone is of no relevance to the need for, or the content of, a plan in that zone. Cost might be taken into account if there were two equally effective means of achieving the objective in view in one particular zone or one local authority area within that zone, but it is illegitimate to decline properly to design or fund the necessary measures in that zone because the benefit to be gained is modest or of limited duration compared with other zones. All that matters is whether such a plan will hasten the achievement of compliance.
78. Furthermore, there is, in respect of the 45 local authorities, no mechanism for enforcing the local plan. On 15 November 2017, the DEFRA Parliamentary Under Secretary of State, Dr Thérèse Coffey, wrote to 33 of the 45 local authorities (those who are not expected to achieve compliance in 2018) encouraging them to bid for the annual air quality grant; stressing the importance of taking action to achieve compliance in the shortest time possible; offering training and materials; and requesting further information on the steps they are taking to achieve compliance. On 19 January 2018, a week before the hearing, a similar letter was sent to the 33. In effect, these local authorities are being urged and encouraged to come up with proposals to improve air quality over the next three years but are not being required to do so.
79. In my judgment, that sort of exhortation is not sufficient. The obligation placed on Member States by Article 23 is to ensure that air quality plans are established; the competent authority in the UK for the purposes of the 2008 Directive is the Secretary of State (see Regulation 3 of the English Regulations); and polite letters from the Government urging additional steps by individual local authorities are not enough. Whilst I see no obligation on the Secretary of State to impose legal directions on local authorities covering every stage in the process of achieving compliance, in my view the failure to make mandatory any step in the case of the 45 means that the Government cannot show either that it is taking steps to "ensure" compliance or, as a result, that compliance is "likely".
80. It follows that the 2017 Plan, in its application to the 45 local authority areas, does not contain measures sufficient to ensure substantive compliance with the 2008 Directive and the English Regulations.
81. Furthermore, each plan must comply with the requirements of the 2008 Directive and the Regulations as to its form. As noted above, Annex XV of the Directive sets out information to be included in local (and other) AQPs. That includes information which identifies the measures being adopted, which sets out a timetable for implementation and provides an estimate of the improvement of air quality planned and the expected time required to attain that objective. Schedule 8 of the English Regulations mirrors those requirements and requires the plans to include details of the measures or objectives adopted, with a description of all the measures set out in the project; the timetables for implementation; an estimate of the improvement of air quality planned and the expected time required to attain those objectives.
82. The local plans produced as part of the 2017 Plan do not meet those requirements. Little time was devoted to the text of the local plans at the hearing but it is apparent that each local plan follows a similar template. After an introduction and general information about the zone (or agglomeration), there is a description of the "overall picture for the 2013 reference year", a section identifying measures that address the exceedances of the NO₂ limit value in the zone and then an analysis of "baseline model projections".
83. In section 4 of each template words to the following effect appear:

"Relevant Local Authority measures within this exceedance situation are listed in Table C.1 (see Annex C). Table C.1 lists measures which a local authority has carried out or is in the process of carrying out, plus additional measures which the local authority is committed to carrying out or is investigating with the expectation of carrying out in the future."

84. A list of measures which have been carried out, are underway, are promised or are being investigated, does not constitute compliance with Annex XV or Schedule 8; it does not amount to a plan describing the measures set out in a project; with timetables for implementation; estimates of the improvement of air quality that will follow and an indication of the expected time required to attain the objectives.
85. Ms Lieven suggests that "feasibility studies" ought to have been required for the 45; Ms Smith counters that these were needed for CAZ but not otherwise. The 2008 Directive and the English Regulations do not specify the development of "feasibility studies", but they do, in my judgment, require the Secretary of State, if he is not to carry out the task himself, to devise some mechanism by which the 45 local authorities can be required to develop plans to address NO₂ exceedances in their areas in a manner that is consistent with the three-fold obligation. "Feasibility studies" is as good a name as any for the first stage of that process.
86. It follows that, as regards those 45 local authority areas, the 2017 Plan does not include the information required by Annex XV of the Directive and Schedule 8 of the English Regulations.
87. As noted above, the circumstances of the 45 local authorities are not homogenous. In particular, 12 are expected to achieve compliance this year. I will hear submissions on relief when this judgment is handed down, but it does not seem to me sensible to require (and I did not understand Ms Lieven to demand) any form of feasibility study in respect of the 12 authorities anticipated to achieve compliance this year. Feasibility studies for measures less complicated than CAZs will undoubtedly take significantly less time than the year or so I understand is required for CAZs. But they will take some time. And thereafter, there will need to be a process by which the outcome of the study is approved and the necessary work commissioned. In those circumstances, it seems to me that the prospect of making any difference to the outcome in these 12 areas is so remote as to make the exercise pointless.
88. For those reasons, and to that extent, this element of the challenge must succeed. Ms Lieven advanced further argument to the same end which it is not strictly necessary for me to address, but in deference to the quality of the argument deployed on this issue, particularly in writing, I set out my conclusions, albeit briefly.

Modelling and Monitoring

89. First, it is said that the 2017 Plan does not sufficiently take into account the results of Local Authority modelling and monitoring in the 45 local authority areas, relying instead on DEFRA's national model. Second, it is argued, the modelling used in the 2017 Plan does not take account of the risk of displacement, i.e. the risk that air quality could be made worse in these 45 local authorities as a result of the displacement of older, more polluting vehicles from the areas that do introduce Charging CAZs. Third, it is argued that the 2017 Plan places reliance on various national measures that it announces, which it appears to assume will have a positive effect on air quality in these Local Authorities, but which have not been modelled. It is said that in consequence the projected compliance of these 45 local authority areas rests on unspecified, un-timetabled measures which have not been modelled. Ms Lieven relies on the witness statements of Dr Claire Holman in support of these arguments.
90. In my judgment none of those points adds anything of substance to the argument.
91. As to the first, I accept the evidence that national monitoring and modelling used for the purposes of the 2017 Plan has been undertaken in accordance with the criteria set out in the Air Quality Directive. I fail to see how that can be criticised on the basis of different results obtained by others that may or may not have been conducted in accordance with the Directive. Further, as Ms Smith contends, "the fact that local modelling may produce different results from those produced by national modelling does not mean that the latter is wrong or "overoptimistic" .
92. As to the second, the evidence demonstrates that the possible effect of displacement was expressly drawn to the attention of local authorities who are to conduct feasibility studies. Both Mr Jackson and Mr Roald Dickens, a senior adviser in DEFRA's Environmental Quality Directorate, make that point. It is right to say that the same point was not made about the 45 local authorities. But that, undoubtedly, is a consequence of the fact that the 45 have not been required to implement feasibility studies to address NO₂ exceedances in their areas in the manner I have now ruled is necessary. I have no doubt that now studies are to be required in the 33 areas, the same point will be made to their local authorities.
93. As to the third, it is plain that the modelling in the 2017 Plan does not rely upon the benefits expected to flow from the non-modelled measures. That means that there are in place additional measures which might reduce exceedances but which are not factored into the calculations. To that extent at least DEFRA's modelling is conservative.
94. I would add that, in my judgment, modelling future compliance with NO₂ limit values is pre-eminently a matter of technical judgement upon which expert opinion is likely to be decisive. DEFRA established an independent panel of experts to provide guidance on this issue. As Ms Smith submits, any challenge to such modelling must show clear legal error or irrationality. I see no such legal error or irrationality here.

The 5 Cities

95. The criticism of the plans for the five cities in the Claimant's Grounds was the lack of any obligation on the cities to comply with the Plan.
96. It was noted that the 2015 AQP had proposed that Charging CAZs would be introduced in Birmingham, Leeds, Nottingham, Derby and Southampton in order to address serious exceedances there; and that the 2017 Plan noted the expectation that they would deliver compliance by 2020. But, it was asserted, no legal requirement to enforce such a timeline was imposed by the 2017 Plan. It was pointed out that the individual AQPs for each of these five cities simply records an expected timeline or an intention for CAZs to be introduced by particular dates, but no obligation to do so.
97. In my judgment, for the reasons set out above in relation to the 45 local authorities, there was some merit in that argument. However, on 19 December 2017, in exercise of the power conferred by s85(5) of the Environment Act 1995 the Secretary of State issued Ministerial Directions to the five cities, (the "Environment Act 1995 (Feasibility Study for Nitrogen Dioxide Compliance) Air Quality Direction 2017"). As submitted by Ms Smith, these impose requirements on the five to submit full business cases to the Secretary of State by 15 September 2018. In my judgment, those Ministerial Directions meet the primary point advanced by Ms Lieven. The critical first step of detailed business cases is now a legal obligation.
98. Ms Lieven complains that there is still no legally mandated timetable for implementation after the business cases are produced. Ms Smith responds that there is no legal obligation to mandate a timetable. She says the timetable is set out in the Plan, and the Ministerial Directions are the first step in ensuring that timetable will be complied with. Further Directions will follow once the business cases have been reviewed.
99. In my view, Ms Smith's analysis on this issue is to be preferred. The Directive and the Regulations require that there must be a timetable, but not that the timetable is itself mandated in law. The Plan as regards the five cities is clear. Paragraph 111 provides:

"The UK government continues to expect local authorities in the five cities named above to deliver their Clean Air Zones by the end of 2019, with a view to achieving statutory NO₂ limit values within the shortest possible time, which the latest assessment indicates will be in 2020."

100. The obligation on the Secretary of State is to ensure that that plan is followed so as to meet the obligations on him imposed by the 2008 Directive and the English Regulations. The issuing of the Ministerial Directions in December 2017 demonstrates how the Secretary of State intends to ensure the Plan will be adopted. Ms Smith made clear that further targeted and tailored Ministerial Directions will be issued in order to require implementation of those measures.
101. In my judgment, the Secretary of State's approach to this issue is a sensible, rational and lawful one. Furthermore, in my view, the clear indication from the Secretary of State as to the next step of the process, is sufficient; were the Secretary of State to fail to act as he has indicated, it is unlikely that this Court would hesitate in requiring him to do so.

Wales

102. Mr Moffett QC told me that, from the outset of these proceedings, the Welsh Ministers have accepted that the Welsh AQP does not satisfy the requirements of either the Directive or the Welsh Regulations and were prepared to give an undertaking that they will correct the position.
103. Accordingly, the only discrete issue that arises in the context of the claim against the Welsh Ministers is that of what remedy, if any, the Court should grant. As to that, it was agreed between Ms Lieven and Mr Moffett that they would seek to agree an appropriate order having seen this judgment in draft. That seemed to me a sensible way to proceed and I will hear submissions from them on relief when this judgment is handed down.

Conclusions

104. For the reasons set out above I conclude that the 2017 Air Quality Plan is unlawful in that:

i) in its application to the 45 local authority areas, it does not contain measures sufficient to ensure substantive compliance with the 2008 Directive and the English Regulations (see paragraph 80);

ii) the 2017 Plan does not include the information required by Annex XV to the Directive and Schedule 8 to the English Regulations, in respect of those same local authority areas (paragraph 86); and

iii) it contains no compliant AQP for Wales (paragraphs 103).

105. I will hear counsel further on the precise details of the relief that is appropriate. But I indicate now that I would be minded:

i) to make a declaration that the 2017 Plan is unlawful in those respects;

ii) to grant a mandatory order requiring the urgent production of a Supplement to the 2017 Plan containing measures sufficient to rectify the deficiencies identified above; and

iii) to direct that the 2017 Plan remains in force whilst the Supplement is produced in order to avoid any delay in its implementation.

106. As indicated above, I will also hear submissions as to the position of the Welsh Ministers.
107. I have given permission to the Defendants to enlarge the group of persons who, upon appropriate undertakings to the Court, may have sight of the embargoed judgment; if a similar application is made by the Claimants I will give it consideration.
108. I end this judgment where I began, by considering the history and significance of this litigation. It is now eight years since compliance with the 2008 Directive should have been achieved. This is the third, unsuccessful, attempt the Government has made at devising an AQP which complies with the Directive and the domestic Regulations. Each successful challenge has been mounted by a small charity, for which the costs of such litigation constitute a significant challenge. In the meanwhile, UK citizens have been exposed to significant health risks.
109. It seems to me that the time has come for the Court to consider exercising a more flexible supervisory jurisdiction in this case than is commonplace. Such an application was made to me when the November 2016 judgment was handed down. I refused it on that occasion, opting for a more conventional form of order. Given present circumstances, however, I would invite submissions from all parties, both in writing and orally, as to whether it would be appropriate for the Court to grant a continuing liberty to apply, so that the Claimant can bring the matter back before the court, in the present proceedings, if there is evidence that either Defendant is falling short in its compliance with the terms of the order of the Court.