England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
ADMINISTRATIVE COURT
THE HONOURABLE MR JUSTICE LEWIS
[2019] EWHC 1997 (ADMIN)

		Royal Courts of Justice Strand, London, WC2A 2LL
		5 August 2020

Covid-19 Protocol: This judgment was handed down remotely by circulation to the parties' representatives by email, release to BAILII and publication on the Courts and Tribunals Judiciary website. The date and time for hand-down is deemed to be 10.30am on 5 August 2020.

Lady Justice Simler:

Introduction

1. The appellant, a trade union representing low-paid workers in the United Kingdom, appeals against part of the order of Lewis J dated 26 July 2019, dismissing a challenge to changes made to the Greater London (Central Zone) Congestion Charging Scheme ("the Scheme") that resulted in the removal of the exemption from payment of the congestion charge for licensed private hire vehicles (referred to as "PHVs" or minicabs as they are more commonly known), unless designated wheelchair accessible, with effect from 8 April 2019.
2. The judicial review claim brought by the appellant (and a number of affected individuals) contended that the removal of the congestion charge exemption unlawfully indirectly discriminated against certain groups of minicab drivers on grounds of race and gender and against some passengers on grounds of disability contrary to s.19 of the Equality Act 2010; and was in breach of articles 8 and 14 of the Human Rights Act 1998. The challenge was rejected on all grounds by Lewis J. The appeal (with permission granted by McCombe LJ on 1 November 2019) relates only to the claims pursued under the Equality Act 2010.
3. The claims proceeded before Lewis J, at least initially, on the accepted basis that the changes to the Scheme involve indirect discrimination against minicab drivers from minority ethnic backgrounds, with the centrally contested question being whether the disproportionate adverse impact was objectively justified. That may be thought unsurprising in circumstances where it was and remains common ground that 94% of licensed minicab drivers in London are from black and minority ethnic backgrounds, with 71% of minicab drivers living in the most deprived areas of London, with earnings of, on average, less than £23,000 (net) per annum. By comparison, 88% of drivers of hackney carriages (known colloquially as black cabs, or taxis) are white and as a group are wholly unaffected by the changes to the Scheme. This significant disparity of impact between black and minority ethnic minicab drivers on low incomes on the one hand and white taxi drivers on the other is stark and has raised legitimate questions about the measure adopted by the Mayor. It has made this appeal particularly troubling.
4. I shall set out the legal and factual background to the impugned decisions before turning to address Lewis J's judgment in the context of the grounds of challenge relied on by the appellant.
5. The appellant is represented by Mr Ben Collins QC, Ms Nadia Motraghi, Ms Nicola Newbegin and Ms Tara O'Halloran. The Mayor and Transport for London have together appeared by Ms Marie Demetriou QC, Mr Malcolm Birdling and Mr David Heaton. I am grateful to all counsel for the excellence of their submissions, both oral and in writing.

The legal and factual background

I. The legal framework

6. The changes to the Scheme were brought into effect by orders made by Transport for London and the Mayor of London under the following framework. The Greater London Authority Act 1999 (the "1999 Act") provides for a Mayor of London and by s.141(1) provides that:

"The Mayor shall develop and implement policies for the promotion and encouragement of safe, integrated, efficient and economic transport facilities and services to, from and within Greater London."

By s.141(3) the transport facilities and services include "those required to meet the needs of persons living or working in, or visiting, Greater London".

7. Those policies are to be included in a transport strategy which the Mayor is under a duty to prepare and publish: s.142 of the 1999 Act. The transport strategy must contain "the Mayor's proposals for the provision of transport which is accessible to persons with mobility problems": s.142(2). In addition, when preparing or revising the policy, the Mayor is required to consult a number of persons or bodies representing the interests of those with mobility problems including the Disabled Persons Transport Advisory Committee.
8. Section 154 of the 1999 Act established Transport for London as a statutory body with functions conferred or imposed on it by the 1999 Act; and requires it to exercise its functions in accordance with guidance or directions issued by the Mayor for the purpose of securing or facilitating the implementation of the transport strategy. In relation to road user charging, s.295 of the 1999 Act empowers Transport for London (among others) to establish and operate schemes for charging road users; and gives effect to Schedule 23. Paragraph 3 of Schedule 23 to the 1999 Act provides:

"A charging scheme may only be made if it appears desirable or expedient for the purpose of directly or indirectly facilitating the achievement of any policies or proposals set out in the Mayor's transport strategy."

9. By paragraph 5, a charging scheme "must be in conformity with the Mayor's transport strategy". Schedule 23, paragraphs 2 and 4, provide that any such charging scheme must be submitted to, and confirmed by the Mayor on behalf of the Greater London Authority.
10. The Scheme created under these powers is set out in the Schedule to the Greater London (Central Zone) Congestion Charging Order 2004 made by Transport for London and approved by the Mayor. The Scheme imposes a charge in respect of an area within Central London referred to as the Congestion Charging Zone ("the CCZ"). Article 4 of the scheme provides for a charge (known as the congestion charge) to be imposed in respect of each charging day on which a vehicle uses or is kept on a designated road, i.e. one in the CCZ, during charging hours. At the material time the charge was specified as £10.50 if a particular payment method (known as Auto Pay) was used, or £11.50 if any other payment method was used. The charges were payable for a vehicle which uses or is kept on a road within the CCZ between 7 am and 6 pm on any day except Saturday, Sunday or certain designated public holidays. (Since the hearing before Lewis J and in response to the Covid 19 pandemic, changes have been made to the level of charge and to charging hours and days, as further indicated below.)
11. Certain vehicles are non-chargeable or exempt from paying the congestion charge. Initially, and prior to the changes with which this appeal is concerned, vehicles exempt from paying the congestion charge included both taxis and minicabs (subject to certain conditions broadly aimed at ensuring that the minicab was being used as such at the relevant time).
12. The instruments effecting the changes in issue on this appeal are the Greater London (Central Zone) Congestion Charging (Variation) Order 2018 made by Transport for London on 29 June 2018 pursuant to paragraph 4(1)(a) of Schedule 23 to the 1999 Act; and the Greater London (Central Zone) Congestion Charging (Variation) Order 2018 and Greater London Low Emission Zone Charging (Variation) (No 2) Order 2018 Instrument of Confirmation 2018, made by the Mayor of London on 19 December 2018, pursuant to the same provision, which confirmed that order. I shall refer to these instruments together as "the measure".

II. The consultation process leading to the measure

13. The process leading to the measure was addressed in detail in the evidence from the Mayor and Transport for London provided to the judge. He summarised it at paragraphs 19 to 45 of the judgment.
14. In overview, in January 2016 the then Mayor asked Transport for London to investigate the potential impacts of removing the exemption of private hire vehicles from liability to pay the congestion charge. A firm of experts in traffic and economic modelling, Cambridge Economic Policy Associates ("CEPA"), were instructed to carry out a preliminary analysis. They conducted further detailed analysis and research in 2017 and early 2018, leading to a report in March 2018 (the "CEPA Report"), to which I shall return below.
15. The Mayor's transport strategy (also published in March 2018) stated that London's streets are "some of the most congested in the world" with extensive impacts on Londoners including harmful pollution, dangers to pedestrians and cyclists from road congestion and serious delay. In relation to road user charging, it noted that although very effective in reducing traffic levels and congestion in the CCZ when the congestion charge was introduced, over time traffic levels had increased (both on weekdays and at weekends). Significantly, it continued,

"the proportion of vehicles in the zone that are subject to the charge continues to reduce as falling numbers of private cars are counterbalanced by increasing numbers of licensed PHVs which are exempt from the charge."

Central London had seen a substantial increase in the number of minicabs with more than 18,000 minicabs entering the CCZ during charging hours each day. It noted the importance of keeping the Scheme under review to ensure that it tackled the congestion in Central London in order to deliver the policies and proposals of the transport strategy. One of those policies (policy 14) was to enhance London's streets and public transport network to make them more accessible and inclusive for disabled people.

16. On 29 June 2018, Transport for London made the variation order and sought confirmation for it from the Mayor. As the judge found, the purpose of the variation order as confirmed and as appeared from its terms was:

"24. …to remove the exemption from the congestion charge previously enjoyed by private hire vehicles save for those which were wheelchair-accessible. Secondly, the underlying aim also appears clearly from the documentation leading to the making of the Order and its subsequent confirmation by the defendant. The aim was to reduce traffic and congestion within central London. Those benefits were to be achieved without reducing the number of wheelchair-accessible private hire vehicles which were considered to provide a means of transport for certain disabled passengers (those whose disability necessitated the use of a wheelchair). The request for a decision noted that although the purpose of the Scheme was to reduce traffic and congestion, there were consequential benefits in air quality resulting from the reduction in the number of vehicles in the CCZ.

25. The request for a decision recorded that the "primary objective of the Congestion Scheme was to reduce traffic and congestion in the CCZ". It noted that the Scheme had initially been very effective in achieving its objective as there had been a marked reduction (30%) on traffic congestion and in circulating traffic (15%). Over time, traffic within the CCZ had increased to levels not seen since before the introduction of the Scheme. One reason for the increase in congestion was the composition in traffic. Another factor was the allocation of road space from traffic use to other uses such as use for cycling or bus lanes.

26. The request noted (footnotes omitted) that:

"London's streets are some of the most congested in the world contributing to poor air quality, delaying vital services and making walking and cycling less attractive options. Without further action, average traffic speeds are forecast to fall across London, with Central London particularly affected. Excess traffic is estimated to be responsible for around 75% of congestion in London so managing demand for road space is crucial.

In addition to inconvenience to the road user, and annual cost of congestion in London is assessed at around £5.5 billion. By 2041, if action is not taken, it will take more than an hour to travel 10km by road in Central London, 15 minutes longer than today. A reduction in traffic of about 10-15% (six million vehicle kilometres per day) is required by 2041 to keep congestion in check, whilst also achieving the aims of the Mayor's Transport Strategy".

27. The request noted that the number of private hire vehicles, and drivers, had increased substantially since the introduction of the Scheme. In 2008/2009, there were around 55,000 licensed private hire drivers and 50,000 licensed vehicles in London and in 2017/2018, there were over 113,000 licensed drivers and over 87,000 licensed vehicles. By contrast, the number of licensed hackney carriages (taxi) drivers and vehicles had remained relatively stable with 24,800 licensed taxi drivers, and 22,300 licensed taxis in 2008/09 compared with 23,826 licensed drivers and 21,026 in 2017/18.

28. At the time that it was decided in 2002 to recommend the exemption of private hire vehicles from the congestion charge, it was estimated that there were around 4,000 private hire vehicles in the CCZ each day during charging hours. By 2017, on an average chargeable day, 18,248 private hire vehicles were seen in the CCZ".

17. A firm of consultants, Mott MacDonald, had been commissioned to undertake an integrated impact assessment of the consequences of removing the exemption of private hire vehicles from the congestion charge (referred to as "the IIA") and, having met and discussed the proposed removal with a number of private hire operators, passenger and trade union groups, they presented a report in July 2018.
18. There was then a consultation exercise on the two main proposed changes to the Scheme as part of the commitment to tackling congestion in Central London and reducing pollution: first, the removal of the exemption from minicabs except for those designated as wheelchair accessible and being used to fulfil a private hire booking; and secondly, replacing the existing Ultra Low Emission Discount with a new phased discount that has progressively tighter criteria and will ultimately cease to be offered in 2025. Other minor proposed changes were also consulted upon but are not relevant to this case. Addison Lee, a large private hire vehicle operator, submitted a report from its consultants, Oxera Consulting LLP ("the Oxera Report"). CEPA responded to that ("the CEPA Reply"). There were also more than 10,000 other consultation responses.
19. Following the consultation, Transport for London produced a document headed Request for Mayoral Decision MD2397 ("the Request") on the proposed changes, which set out the facts and advice to the Mayor, together with a number of supporting documents. These included a report to the Mayor on the consultation (referred to as "the Consultation Report") dated December 2018, which described the proposals, summarised the consultation results and the responses to issues raised. At paragraph 2.5 it summarised the key findings made by the IIA carried out by Mott MacDonald in relation to the proposed changes, indicating that the IIA was published as part of the consultation materials and was attached to the Consultation Report as part of appendix B. Also attached as appendices to the Consultation Report were the CEPA Report, the Oxera Report, the CEPA Reply and a summary of the consultation responses.
20. On 17 December 2018 the Mayor decided to confirm the order by making the confirmation order on 19 December 2018.

III. The expert reports

21. Lewis J summarised the terms and effect of the CEPA Report as follows:

"29. The CEPA report forecast that the removal of the exemption from private hire vehicles (other than wheelchair-accessible vehicles) could result in a 45% reduction in the number of private hire vehicles entering the CCZ each day (although those which entered the CCZ may remain there and carry out more journeys within the CCZ). It forecast that there could be a 6% reduction in the number of private hire vehicles in the CCZ overall. That amounted to a forecast reduction of 1% of traffic in the CCZ overall (i.e. private hire vehicles and other vehicles).

30. Reading the CEPA Report, and the request for a decision, it seems that the forecast was based on a certain number of assumptions. It seems that a reduction in congestion and traffic will result only if either fewer journeys are made into the CCZ or if a smaller number of vehicles carry out more journeys in the CCZ. If, for example, passengers are required to bear all or part of the cost of the congestion charge but considered that to be too expensive, they might switch to (cheaper) public transport or decline to travel into the CCZ. In relation to the second possibility, if, to use a theoretical and very simplified example, there were 20 private hire vehicles each completing one journey into the CCZ prior the congestion charge, but changes in behaviour mean that 10 no longer travel into the CCZ and the work into, within and out of the CCZ is carried out by the other 10 vehicles, there may be a similar amount of journeys made but the number of vehicles involved in making those journeys is reduced. That may result in fewer vehicles in the CCZ. Consequently, there could in theory be a reduction in congestion within the CCZ.

31. On the first possibility, the response of passengers to increases in price due to part or all of the congestion charge being passed on to the passenger, CEPA forecast that customers were likely to be fairly price sensitive although this was likely to be less so in the case of small operators with localised markets and more loyal customers. CEPA forecast, however, that some customers would switch from one private hire vehicle operator to an alternative operator in order to find a lower price. They also considered that some of the customers who switched would change from private hire vehicles to taxis. They did not forecast that most passengers who did change their behaviour would switch away from private vehicles or taxis to public transport such as buses or the tube. Rather, CEPA said that it would "assume most customers would remain in the PHV/taxi sector with 'switching' customers allocated to operators with low fares and a large existing CCZ presence". That meant that some journeys would not be made by private hire vehicle (or taxi) and would either be made by public transport or not made at all. That would result in fewer vehicles, i.e. those that would otherwise have been used for such journeys, in the CCZ.

32. CEPA then considered changes in behaviour by private hire vehicle operators and the extent to which changes in behaviour could reduce traffic and congestion. CEPA noted that specialisation could occur when journeys were undertaken with fewer vehicles entering the CCZ, for example, by designating certain private hire vehicles for work within the CCZ or allocating work to private hire vehicles already within the CCZ. CEPA assumed that only the largest operators (in effect, two of the current operators) would be able to specialise in that way. Furthermore, drivers could also choose to spend more time within the CCZ, having paid the congestion charge.

33. Given all the assumptions made, CEPA forecast that private hire vehicle traffic within the CCZ could reduce by 6%. That could amount to a decrease in 1% in the overall traffic, i.e. the number of vehicles, in the CCZ in a year. It is not possible from the report to determine what proportion of the 6% forecast decline in traffic resulted from the reduction in demand from passengers for private hire vehicle and what proportion resulted from specialisation, that is the same number of journeys being undertaken in the CCZ by a lower number of private hire vehicles than was previously the case.

34. The report noted that the greatest impact of the change would be on small operators as they would be likely to be less able to accommodate the increased charge".

22. The judge then dealt with the Oxera Report and the CEPA Reply in the following terms:

"35. The Oxera Report commissioned by Addison Lee took issue with the CEPA assessment and the assumption that there would be specialisation in the use of private hire vehicles within the CCZ. CEPA considered the matter again. In their reply, they set out the rationale for the specialisation assumption and their belief that there was some scope for specialisation. They indicated that the major driver, by which I understand they mean cause, of the predicted reduction in traffic would be the reduction in demand for private hire vehicles.

36. In other words, it seems that the bulk of the anticipated reduction of 6% private hire vehicles would result from a reduction in the amount of passenger journeys. That is, there will be less work available for private hire vehicles. In addition, some of the journeys still undertaken by vehicles will be undertaken by taxis, not private hire vehicles as some passengers will switch to taxis.

37. The issue is dealt with in the request for a decision in the following terms:

"2.15 The CEPA Report forecasts that the removal of the PHV exemption would reduce traffic and congestion in the zone. In summary, they forecast that during charging hours in the Congestion Charging zone there would likely be:

45% reduction in unique PHV entries;

6% reduction in PHV traffic; and

1% reduction in traffic overall.

(CEPA who forecasted these figures have stated that they represent broad estimates only but given the conservative approach they adopted, CEPA consider the 1% figure is at "the lower end of the range" (see page 5 of CEPA's further response dated 9 November 2018 (Appendix E to the Report to the Mayor which is attached at Appendix 2)).

2.16 CEPA's forecast of a 45% reduction in the number of unique entries by PHVs into the zone during charging hours is based on the assumption that operators with larger fleets will distribute their bookings to minimise the number of PHVs needing to enter the CCZ. Doing so would mean that a smaller number of vehicles specialise in taking bookings within the zone, potentially undertaking more trips in the zone than they would have previously. The greatest impact on congestion will result from the expected lower demand for PHVs in the CCZ during charging hours as a consequence of the price per journey increasing to reflect the cost of the Congestion Charge.

2.17 Some have criticised CEPA's forecasts, in particular the Addison Lee Group who have submitted an independent report by Oxera, which contended that specialisation was a flawed concept and therefore the predicted traffic reduction of 1% was also unlikely to be realised. In response, TfL commissioned further work from CEPA to consider Oxera's report. The further work by CEPA (Appendix E to the Report to the Mayor at Appendix 2) states that their view remains that some specialisation is likely to occur as a response to competitive pressure. It also clearly acknowledges that there are uncertainties with regard to the scale of specialisation; however, the impact on traffic is not dependent on their judgement of specialisation. CEPA also explained that the 1% reduction in traffic is a conservative estimate. TfL support CEPA's analysis and note that the reduction in traffic is more closely tied to the demand response than specialisation. TfL also note that a 1% reduction in traffic in the zone is not an insignificant benefit where the potential for more radical change (during charging hours) is very limited, but congestion is still very high.

2.18 Although the purpose of the Congestion Charging Scheme is to reduce traffic and congestion, there have always been consequential improvements in air quality from doing so. Removing the exemption for PHVs should reduce the number of vehicles in the zone and, therefore, help to improve air quality. The introduction of the CVD may further incentivise PHV drivers continuing to drive in the CCZ to do so in the cleanest possible vehicle. These will complement other initiatives including the introduction of the ULEZ Scheme in April 2019.""

23. So far as the equality impacts identified in the Mott MacDonald report are concerned, Lewis J referred to the Request. In the advice to the Mayor, at section 4, it set out the public sector equality duty which applied to the Mayor's decision, stating that the equality impact assessment in respect of the removal of the PHV exemption anticipated "some long-term minor impacts, both negative and positive, to some groups with protected characteristics" as follows:

"BAME PHV drivers


• Around 94% of all PHV drivers are from a BAME background so they will be disproportionately impacted by the removal of the exemption;

• Increased professional costs as a consequence of having to pay the Congestion Charge will be incurred. Those PHV drivers who enter the CCZ every day during charging hours could expect to pay around £230 a month (assuming a 22-working day month and use of Auto Pay). In cases where a driver would need to absorb all costs, and travel in the zone every day, the impact would be at its greatest. This scenario is unlikely to be typical, except in cases of specialisation (which itself implies that the business model is set up to absorb the costs beyond just the driver);

• Overall, the impact is assessed as a minor adverse one because the distribution and scale of the impact is considered to be low. Not all drivers will regularly enter the CCZ in charging hours. The taxi and private hire driver diary survey undertaken by Steer Davies Gleave in 2017 indicates 33% of the sampled PHV drivers made journeys into the CCZ in charging hours, while 23% of trips involved travel to, from or within the CCZ. This data would suggest that there is a fairly large number of PHVs which do not regularly enter the CCZ and so would not see a substantial increase in operating costs through paying the Charge. As suggested in the CEPA Report, some operators may take on the costs themselves or choose to pass the cost on to passengers; and

• Sensitivity to the impact was assessed as low because the overall financial costs will be reduced if the payment constitutes a tax deductible expense, drivers qualify for a 100% discount such as the CVD or they are able to spread the cost over multiple trips.

Part-time female PHV drivers

• A higher proportion of women across all industries tend to work part-time as compared to men (42% of women versus 13% of men). Part-time PHV drivers will be less able to spread the cost of the Congestion Charge across a number of journeys. As women can be assumed to be more likely to work as part-time PHV drivers, they will be disproportionately affected.

• However, this impact was assessed as being "very low" as women make up less than 2% of PHV drivers in London, of which not all will work part-time, or in the CCZ during charging hours. Notwithstanding the scale of the impact, for those women who fall within the 2% and work day time weekday shifts in the CCZ, sensitivity to increased professional costs may be high.

Drivers from deprived communities

Within London, 71% of PHV drivers live in areas which are within the most deprived and second most deprived quintiles, as defined by the index of multiple deprivation … The reduction in income that may be faced by drivers who are required to pay the congestion charge may be prohibitive in terms of some PHV drivers staying within the profession or entering it. Given that this trade offers an employment stream for communities in deprived areas, it can be expected that the increased costs as a result of paying the congestion charge may impact disproportionately on communities in London where there are areas of high deprivation.

The impact is considered to be a minor adverse one for the same reasons stated in respect of BAME drivers: not all drivers will regularly enter the CCZ in charging hours and some operators may take on the costs or will pass the cost onto passengers…

It is, therefore, likely that a high proportion of PHV drivers would continue within the profession."

24. As the judge recorded, the Mott MacDonald report also dealt with the impact on passengers, including disabled passengers. It noted that those with mobility problems use private hire vehicles more frequently than those without (8% of disabled Londoners use them once a week compared with 6% of non-disabled Londoners). It noted that the impact was only relevant to those travelling within the CCZ during charging hours with the additional cost being nil if absorbed by the driver or operator, or minimal if spread by the driver or operator over a number of hires. It thought that the most likely scenario was that the charge would be passed on to customers who would pay around £1 to £2 a trip. The request recognised that that sum may not be insubstantial for people such as the disabled, people on low income or women, but noted that there would be opportunities to minimise the impact. Those included the continuation of the exemption for wheelchair-accessible minicabs. There were other mitigating opportunities including, for those disabled people with Blue Badges, the ability to nominate the vehicle for a journey and thereby be exempt from the congestion charge. There were also subsidised travel schemes for taxis and minicabs that certain passengers could use.
25. The IIA and the Consultation Report expanded on the potential impacts and the consequences of the proposed removal of the exemption from PHVs. In terms of negative impacts, the Consultation Report said:

"Negative impacts

2.5.13 This proposal may put pressure on earnings for PHV operators and drivers. This could result in negative health outcomes for individuals. It may be difficult for some individuals to cover these costs and as such the removal of the exemption may lead to stress related and mental health issues for PHV drivers. It may also impact on physical health as a result of potential longer working hours.

How sensitive PHV drivers are to this impact will depend upon whether they meet the criteria for alternative discounts and exemptions, whether they are able to pass all or some of the cost onto passengers, whether they can share all or some of the cost with operators and whether they can adapt their behaviour to operate in the CCZ outside charging hours or outside the CCZ. Additionally, the IIA notes that the cost of the Congestion Charge may be tax deductible as a business cost (for drivers and/or operators). In cases where a driver would need to absorb all costs, and travel in the zone every day, the impact would be at its greatest (around £230 a month assuming a 22-working day month and use of Auto Pay).

This scenario is unlikely to be typical, except in cases of specialisation (which itself implies that the business model is set up to absorb the costs beyond just the driver). And as stated at the beginning of this chapter, the proposal may only affect a relatively small proportion of PHV drivers as two thirds of PHV drivers do not enter the CCZ in charging hours. The overall impact was assessed as minor adverse.

2.5.14 This proposal may lead to some smaller PHV operators experiencing a rise in price per trip and potentially a reduction in demand for their services. This may lead to poor health and wellbeing outcomes for operators. The sensitivity of operators to this impact will depend on a number of scenarios including whether drivers will absorb some or all of the cost, if drivers are eligible for alternative discounts, whether drivers frequently enter the CCZ and the ability to spread costs over multiple trips. This was assessed as a minor adverse effect.

2.5.15 The removal of the PHV exemption may also limit the ability of older or disabled passengers to access essential services related to their health and wellbeing. Although designated wheelchair accessible PHVs will remain exempt, disabled passengers who do not use a wheelchair could see an increase in fares of around £1-2 for trips in the CCZ, depending on how the cost is passed on, unless they are eligible for another discount or exemption (e.g. the Blue Badge discount). This was assessed as a minor adverse effect.

2.5.16 It is recognised that any increase in fares may not be an insubstantial sum for these categories of people. However, there may be opportunities to avoid increased fares or minimise the impact of them. Older and disabled passengers may also be eligible to use Taxicard services to access essential services related to their health and wellbeing. Black cabs are used to carry out around 90 per cent of Taxicard journeys. Capped fares for Taxicard journeys in black cabs are expected to come into effect from 1 January 2019. See paragraph 2.5.7 – 2.5.9 below for more information as to mitigation."

26. The Consultation Report also noted the negative equalities impact of the proposed removal of the exemption from PHVs, as follows:

"Negative impacts

2.5.33 This proposal may negatively impact PHV drivers, particularly those that operate regularly in the zone during charging hours as they find their costs increase and incomes reduce as they cover some or all of the cost of the charge. As the majority of PHV drivers (around 94 per cent) are from Black, Asian and minority ethnic backgrounds (BAME) and many are from deprived areas, there is a disproportionate impact on these groups. There will also be a very low impact on part-time female PHV drivers (although women make up less than two per cent of PHV drivers in London). The impact on BAME PHV drivers and female PHV drivers was assessed as minor adverse.

2.5.34 The overall financial costs will be reduced if the payment constitutes a tax-deductible expense, or drivers are able to spread the cost over multiple trips. It could be neutralised if vehicles qualify for other 100 per cent discounts or exemptions such as the CVD.

2.5.35 Negative equality impacts on passengers are most likely to affect those on low incomes, female and disabled passengers who are more frequent users of PHVs and would be disproportionately impacted if fares increase or PHV availability declines. The impact is only relevant to those passengers who wish to travel within the CCZ during charging hours with the additional costs nil (if absorbed by the driver or operator) or minimal (if spread by the driver/ operator over several hires)."

27. The Consultation Report also discussed the negative equality impacts on passengers including disabled passengers who were more frequent users of private hire vehicles and would be disproportionately affected if fares increased or the availability of private hire vehicles declined. It noted matters that may mitigate that impact.
28. The Consultation Report noted that the impact on private hire vehicle drivers would only be experienced by those who drive into the CCZ during charging hours. Only about 33% of drivers did so. 67% of private hire vehicle drivers did not. It noted that passengers would experience reduced choice if fewer operators offered a service into the CCZ and that may also increase the cost of fares if operators passed on the cost to passengers.
29. The material available to the Mayor made clear that there was no proposal to remove the exemption from the congestion charge for taxis. It noted that the number of licensed taxi cabs and licensed taxi drivers had remained static or gradually declined over time. The Request noted that there was believed to be "social value" in retaining the exemption for taxis given that in certain circumstances they represented the only method of transport available to wheelchair using disabled persons. The Request noted that taxis are legally required to be wheelchair-accessible (whereas private hire vehicles are not and only 525 PHVs – less than 1% of the total number of licensed PHVs – are wheelchair-accessible). Taxis are also required to provide a range of other features which made them better placed to meet the needs of certain categories of passengers. The Request noted the vital role played by taxis in the transportation of disabled passengers in central London and considered that the exemption for taxis should remain. The Request said that:

"For these same reasons, the proposals safeguard the exemption for the small number of PHVs which are designated wheelchair accessible."

The appeal

30. Lewis J held that the aim of the measure was to reduce traffic and congestion within the CCZ without reducing the number of wheelchair-accessible minicabs available. That was a legitimate one, and was adopted as a measure of economic, social and environmental policy. He held that the measure adopted corresponded to a real need to reduce traffic congestion, was appropriate and suitable to achieve that aim, and no other alternative measures that would realistically achieve the aim to the same extent had been shown. As to proportionality, Lewis J held that the measure was a proportionate means of achieving the legitimate aim of reducing traffic and congestion within Central London so that any indirect discrimination was not unlawful.
31. Each of the conclusions reached by Lewis J is challenged by the appellant. I will deal with his reasoning for each conclusion when I come to address the relevant ground of appeal below. The following grounds of appeal are advanced by the appellant:

i) The judge erred in finding that removal of the congestion charge exemption from over 99% of minicab drivers was "appropriate" with regard to the legitimate aim relied upon, by failing to carry out a proper evaluation of the effectiveness of the measure.

ii) The judge erred in concluding that the aim properly amounted to a legitimate aim for the purposes of the Equality Act 2010.

iii) The judge erred in finding that removal of the exemption from over 99% of licensed minicab drivers was reasonably necessary to achieve the aim sought.

iv) The judge erred in his approach to proportionality by not considering whether the measure itself (the provision, criterion or practice, or "PCP") as opposed to its impact, was justified.

v) The judge erred in his approach to s.19(2)(b) Equality Act 2010 and the nature of the comparison between those with the relevant protected characteristic and those without it.

vi) The judge failed to carry out a proper proportionality exercise.

vii) The judge erred in law in giving improper weight to the fact the decision maker was the state.

32. There is also a Respondent's Notice that arises only if the appellant succeeds on the appeal. In that event the Mayor contends that s.19(2)(b) Equality Act 2010 is not in fact satisfied in relation to minicab drivers, so the measure cannot constitute unjustified indirect race discrimination. Although as indicated above, there was a concession that s.19(2)(b) was satisfied in this case, the respondent sought to withdraw the concession in the course of the hearing before Lewis J. The judge said he would have been minded to permit the Mayor to withdraw the concession but in the event decided that he did not need to address this question in light of his other conclusions. In short summary on this point, the Mayor (supported by Transport for London) contends that the appropriate pool of comparators for the purposes of s.19(2)(b) Equality Act 2010 is all minicab drivers, rather than all minicab and taxi drivers, in London. They submit that there are material differences between taxi and minicab drivers; and that minicab drivers from minority ethnic backgrounds are not put at any particular disadvantage relative to white minicab drivers because all minicab drivers have to pay the congestion charge and are similarly disadvantaged in that regard. The appellant resists these contentions, arguing that the original concession was correctly made and that the pool for comparison purposes is all minicab and taxi drivers in London.
33. As just indicated, it was accepted until partway through the hearing before Lewis J that the measure put both drivers from minority ethnic backgrounds and female drivers at a particular disadvantage so that the sole issue was whether the Mayor could show it was a proportionate means of achieving a legitimate aim within the meaning of s.19(2)(d) Equality Act 2010. The judge proceeded on that assumption and for the purposes of the appeal I proceed on that basis too.
34. Before addressing the specific grounds of appeal, I set out the definition of indirect discrimination under s. 19 of the Equality Act 2010 which prohibits the application of a neutrally expressed "provision, criterion or practice" (a "PCP") to a group of people that has a disparate adverse impact on members of the group with one or more particular protected characteristics and cannot be justified as a proportionate means of achieving a legitimate aim.
35. Indirect discrimination under s.19 of the Equality Act is defined as follows:

"(1) A person (A) discriminates against another (B) if A applies to B a provision, criterion or practice which is discriminatory in relation to a relevant protected characteristic of B's.

(2) For the purposes of subsection (1), a provision, criterion or practice is discriminatory in relation to a relevant protected characteristic of B's if—

(a) A applies, or would apply, it to persons with whom B does not share the characteristic,

(b) it puts, or would put, persons with whom B shares the characteristic at a particular disadvantage when compared with persons with whom B does not share it,

(c) it puts, or would put, B at that disadvantage, and

(d) A cannot show it to be a proportionate means of achieving a legitimate aim.

(3) The relevant protected characteristics are –

…. disability; … race; …sex; …."

36. It is not in dispute that s.19 applies to the exercise of public functions by virtue of s. 29 Equality Act 2010.
37. The proper approach to the question at s.19(2)(d), whether the measure is a proportionate means of achieving a legitimate aim, was largely common ground before the judge and remains so before this court. The relevant principles were set out by Lewis J by reference to Lockwood v Department of Work and Pensions [2013] EWCA Civ 1195, [2014] ICR 1257 (recognising the employment context would require suitable modification to reflect decision making in a public law context) as follows:

"10. The legal principles with regard to justification are not in dispute and can be summarised as follows:

(1) The burden of proof is on the respondent to establish justification: see Starmer v British Airways [2005] IRLR 862 at [31].

(2) The classic test was set out in Bilka-Kaufhas GmbH v Weber Von Hartz (Case 170/84) [1984] IRLR 317 in the context of indirect sex discrimination. The ECJ said that the court or tribunal must be satisfied that the measures must "correspond to a real need … are appropriate with a view to achieving the objectives pursued and are necessary to that end" (para 36). This involves the application of the proportionality principle, which is the language used in regulation 3 itself. It has subsequently been emphasised that the reference to "necessary" means "reasonably necessary": see Rainey v Greater Glasgow Health Board (HL) [1987] ICR 129 per Lord Keith of Kinkel at pp 142–143.

(3) The principle of proportionality requires an objective balance to be struck between the discriminatory effect of the measure and the needs of the undertaking. The more serious the disparate adverse impact, the more cogent must be the justification for it: Hardys & Hansons plc v Lax [2005] IRLR 726 per Pill LJ at paras 19–34, Thomas LJ at 54–55 and Gage LJ at 60.

(4) It is for the employment tribunal to weigh the reasonable needs of the undertaking against the discriminatory effect of the employer's measure and to make its own assessment of whether the former outweigh the latter. There is no "range of reasonable response" test in this context: Hardys & Hansons plc v Lax [2005] IRLR 726, CA."

38. It is well established that it is for the court to conduct an objective assessment of the evidence for itself in order to decide whether an impugned measure is a proportionate means of achieving a legitimate aim, rather than merely exercising a review jurisdiction. There must be a critical and thorough evaluation of the evidence by the first instance judge. The appellate court does not re-perform that assessment (save where relevant new evidence is admitted) but considers whether the reasoning of the judge below was justified: see R (Friends of Antique Cultural Treasures Ltd) v Secretary of State for Department of Environment, Food and Rural Affairs [2020] EWCA Civ 649 at [6].

Grounds 1 and 2: Appropriateness and 'no reduction'

39. I take these grounds together because there is some overlap and they are both directed at the reasoning and conclusions reached by the judge at paragraphs 74 and 75 of the judgment. Under these grounds the appellant challenges Lewis J's conclusion that the removal of the congestion charge exemption from over 99% of minicab drivers was appropriate to secure the aim of reducing traffic and congestion within the CCZ without reducing the number of wheelchair accessible vehicles (considered to provide a means of transport for those disabled passengers requiring or choosing to travel in a wheelchair) for private hire. The judge held as follows:

"74. The measure adopted by the defendant does correspond to a real need. The need is to reduce the number of vehicles within the CCZ. The removal of the exemption from the congestion charge for private hire vehicles does reflect, or correspond, to a real need. The number of private hire vehicles had increased substantially since the Scheme was introduced in 2003. The number of private hire vehicles had increased from around 50,000 to over 87,000 and the number of licensed private hire vehicle drivers has more than doubled from around 55,000 to over 113,000 between 2008/2009 and 2017/2018. There was no similar increase in taxis or taxi drivers. The forecast was that the removal of the exemption could lead to a reduction of 6% in the number of private hire vehicles in the CCZ (which would amount to a reduction of 1% in traffic overall). The measure would, therefore, address the need to reduce the number of vehicles in the CCZ. Furthermore, where a measure is intended to operate, at least in part, by seeking to change the behaviour of individuals (here operators and drivers of private hire vehicles, and passengers) it is reasonable to rely on forecasts of the likely change. At the time that the measure was adopted, the forecast was of a reduction. The defendant was entitled to rely on those forecasts. It is also permissible to have regard to subsequent events to determine whether the forecast changes have materialised. As it happens, in the relatively short time since the amendments to the Scheme came into force, and bearing in mind there may be other factors at play, the evidence is that the reduction in traffic is broadly in line with the forecast. The evidence of Ms Calderato, in her second witness statement, is that private hire vehicles are making between approximately 3,000 and 6,800 fewer entries into the CCZ.

75. The removal of the exemption from private hire vehicles (other than the small number of wheelchair-accessible vehicles) was an appropriate and suitable method of reducing the number of vehicles in the CCZ. It could result either in fewer journeys into the CCZ or a more rational use of the road space within the CCZ with a smaller number of private hire vehicles performing the number of journeys previously undertaken by a larger number of such vehicles (and reducing the number of vehicles travelling empty within the CCZ). The method adopted was reasonably necessary for that end. Unless steps were taken to reduce the number of private hire vehicles in the CCZ there would remain a real problem of a large number of private hire vehicles driving into the CCZ. Taking steps by removing the exemption from congestion charge would have an impact on that as it was forecast to result in fewer trips into, and better use of private hire vehicles within, the CCZ."

40. The appellant's essential challenge to these conclusions by reference to ground one is that there was a failure by the judge to engage with the limited positive benefits of the measure: the 1% reduction in traffic overall was so small and was in any event, unreliable, being based on forecasts and dependent on behavioural changes, including specialisation by private hire operators that might not materialise. Furthermore, the evidence showed that 19% of minicab passengers are registered disabled compared with 14% of taxi passengers, yet there was a failure to consider, quantify and assess the needs of those disabled passengers who prefer to travel whilst not sitting in a wheelchair or whose mobility impairment or disability does not require them to travel while sitting in a wheelchair, but who have a greater need for travel by minicab.
41. I do not accept that Lewis J made any error in coming to the conclusions he did. In my judgment, he was entitled on the evidence before him to do so for the reasons that follow.
42. First, it is clear from the judgment read as a whole that Lewis J did engage closely with the material available to the Mayor when making his decision. As indicated above, he summarised that material at length at paragraphs 29 to 45 and plainly understood its import. He set out the nature and seriousness of the congestion problem as discussed in the Mayor's transport strategy and in the Consultation Report. He expressly recognised that the CEPA forecast of a 6% reduction in the number of private hire vehicles in the CCZ amounting to a forecast reduction of 1% of traffic in the CCZ overall, was based on assumptions made by CEPA: see to that effect his observation at paragraph 30, "Reading the CEPA Report, and the request for a decision, it seems that the forecast was based on a certain number of assumptions". These assumptions were identified by him from the material as including the likely behaviour of passengers in response to increased prices due to part or all of the congestion charge being passed on to them, resulting in lower demand for minicabs and fewer such journeys; and changes in behaviour by minicab drivers and operators, again resulting in fewer journeys into the CCZ.
43. Secondly, Lewis J was plainly alive to the criticisms made in the Oxera Report and in particular, that Oxera took issue with the 1% assessment as unreliable and unlikely to be realised, and criticised the assumption that there would be specialisation in the use of private hire vehicles within the CCZ. These criticisms, together with the CEPA Reply to them are adequately reflected in the judgment and I reject Mr Collins QC's criticisms to the contrary effect. It is not necessary for a judge to conduct a line by line analysis of all the evidence available, nor is there any duty on a judge, in giving reasons, to deal with every argument presented by counsel in support of his or her case. The judge's function is to reach conclusions and give reasons to support those conclusions. That is what Lewis J did.
44. It is implicit in the conclusions reached that the judge accepted CEPA's forecast reduction of 1% overall as a conservative estimate that was not dependent on specialisation by PHV operators, but was more closely tied to the demand response. Having done so he expressly accepted that it was reasonable for the Mayor and Transport for London to rely on forecasts of the reduction in traffic overall in circumstances where the measure was intended to operate by seeking to change behaviours of operators, minicab drivers and passengers for the future. Finally, although considered in isolation divorced from its relevant context, a 1% reduction overall may be regarded by the appellant as very small, it is clear that Transport for London viewed a 1% reduction in traffic in the CCZ as meaningful, describing it as a not "insignificant benefit where the potential for more radical change (during charging hours) was very limited but congestion is still increasing". Moreover, the forecast of a 1% reduction must be set against the inevitability of future traffic increases overall caused by rising numbers of minicabs in the CCZ if nothing were done. The judge concluded that the measure corresponded to a real need to reduce traffic and congestion and on the evidence, he was entitled to do so.
45. Thirdly, the judge was entitled to consider the extent to which evidence of subsequent events reinforced or supported the changes forecast by CEPA, and the criticisms of his conclusion that they did are misplaced. The judge referred to the evidence of Ms Calderato, Head of Transport Strategy and Planning for Transport for London, of a sustained decrease of between 3000 and 6800 (approximately) entries into the CCZ by minicabs in the short period after the changes were implemented. Ms Calderato acknowledged the fact that the simultaneous introduction of the ULEZ charge might be contributing to the reduction and recognised that the data could not determine the relative contribution of any one factor over another to the reduction in unique PHV entries into the zone. However she was able to identify the percentage of minicabs complying with ULEZ standards as at 31 May 2019 (72.6%) and who were therefore not required to pay the ULEZ charge when entering the zone and inferred as a consequence that the reduction in unique PHV entries into the CCZ was "in large part due to" the measure. She concluded that "the reduction remains within or just below the range CEPA predicted on a monthly basis". The judge expressly recognised the limitations of the evidence about subsequent events and the fact that other factors might be at play, but in light of the evidence was entitled to reach the conclusions he did in this regard.
46. Fourthly, to the extent that the judge is criticised by the appellant for making factual assumptions about the behaviour of minicab drivers circulating without passengers in the CCZ, I regard these criticisms as misplaced. There was evidence from Ms Calderato who referred to surveys conducted by Transport for London showing this to be the case. Nor did the judge assume, as the appellant suggests, that minicab drivers would be able to recall every postcode in London in order to avoid driving into the zone by refusing to accept jobs with postcodes in the CCZ. Rather, he identified Transport for London's expectation that drivers would become more adept over time at using the destination postcode system to avoid journeys into the CCZ; or alternatively if that did not occur, that Uber would adapt the system in order to enable drivers to choose whether or not to enter the zone.
47. Finally, under both the first and second grounds Mr Collins submits that the measure could not be appropriate or have a legitimate aim, because the Mayor and the judge failed to have regard to the wider needs of disabled passengers as opposed to simply those requiring wheelchair accessible vehicles as narrowly defined, and thereby favoured one group of disabled passengers over other groups of disabled passengers who are disadvantaged by it. The appellant contends that this point has all the more force in circumstances where there is (and was) no evidence (by way of data or otherwise) that the availability of wheelchair accessible transport provision before the changes represented an oversupply, an undersupply or the right level of supply. Mr Collins relies on the opinions of the Advocates General in Dansk Jurist – og Okonomforbund v Inderigs-og Sundhedsministereiet [2014] ICR 1 at [AG74] and VL 18 June 2020 Case C-16/19 at [AG27 and AG41] by analogy, to submit that where a measure favours one group of wheelchair users while at the same time being to the detriment of a wider group of disabled people it is inconsistent and incoherent, and cannot in principle have a legitimate aim. The judge was accordingly wrong to conclude that it was.
48. Before dealing with these points, I need to address a number of concerns arising out of the way in which the aims of this measure were stated by the Mayor and Transport for London. In the course of the hearing all members of the court were troubled by what appeared to be the apparent insertion of the requirement that there be no reduction of wheelchair accessible vehicles into the principal aim of reducing traffic and congestion.
49. We had two concerns. First, we wondered whether the maintenance of the existing level of wheelchair accessible vehicles was really an aim at all, or rather whether it was, in reality, a fixed element of the measure. If that were, the case, the evaluation of whether the measure was appropriate might have been skewed. Put simply, we wondered whether stating the maintenance of the existing level of wheelchair accessible vehicles as a second part of the aim could be regarded as a device designed to achieve a pre-determined outcome whereby taxi drivers were preferred over PHV drivers. Secondly, it seemed to us that the first aim of reducing congestion could be regarded as having been inaccurately stated: our concern was that the real aim was simply to reduce or stem the increase in the number of minicabs in the CCZ. As Singh LJ observed in the course of argument, if on a true analysis, the true aim was simply to reduce the number of minicabs, there would be real concern about the legitimacy of such an aim bearing in mind the close correlation between minicab drivers and people from black and minority ethnic communities on the one hand, and the stark contrast with the predominantly white taxi driver workforce. These matters were raised with the parties who were given time and an opportunity to respond to our concerns.
50. Ultimately I am persuaded by Ms Demetriou QC that whether described as part of the means or as an aim in itself, maintaining the current levels of wheelchair accessible vehicles for disabled passengers, in the form of taxis, is both reasonable and legitimate for the reasons given below, and was not a device to protect taxis at the expense of minicab drivers. The question of removing the congestion charge exemption for taxis was raised during the consultation process but was rejected for a number of legitimate reasons. First, taxis are subject to different regulatory rules that legally oblige them to be wheelchair accessible and to provide a range of other accessibility features for disabled passengers. Secondly, they are compelled to accept any hire within a 6 mile radius of Charing Cross of up to one hour in duration or 12 miles long. This means they must accept a hire where the destination is the CCZ and refusing to do so amounts to a potential offence. Thirdly, they are expected to take the shortest, most direct route to fulfil a hire. Fourthly, there is no ability for taxi drivers to set their own fares which are regulated by Transport for London. Taxis could not therefore recoup the charge from passengers under the current regulations. Accordingly, a different scheme involving removing the exemption for taxis would have required changes to the regulations governing taxis and would, inevitably, have involved a far more complicated scheme. As Ms Demetriou was entitled to emphasise, it is legitimate for a decision maker to consider the ease with which a measure can be administered and its simplicity; and open to a decision maker to reject a potential alternative as too complicated.
51. So far as the first aim of reducing traffic congestion is concerned, Ms Demetriou concedes that if the aim was not to reduce congestion but was in fact simply to reduce the number of minicabs, it would not be legitimate. There was and is no power to cap the number of minicabs licensed by Transport for London and to seek to achieve this result by means of the congestion charging scheme would be ultra vires its purpose. However, for the reasons which follow, I accept that the reduction in the number of minicabs in the CCZ was not an aim in itself for some extraneous reason unrelated to congestion but was the means by which the Mayor's aim of reducing congestion was and is to be achieved. Transport for London's advice in the Request for a Mayoral decision set out an analysis of the way in which exemptions from payment of the congestion charge were having a direct impact on the effectiveness of the Scheme. The evidence demonstrated that the number of cars in the CCZ was reducing, the number of taxis was stable, but the number of minicabs, all exempt from payment of the charge, was increasing to a degree that was never envisaged. In those circumstances, there were sound and legitimate factual reasons for concluding that the most effective mechanism for reducing congestion was to reduce or stem the number of minicabs entering the CCZ. While it is true that the measure removes the PHV exemption from the congestion charge (while leaving the taxi and wheelchair accessible minicab exemptions in place), that is only because on the analysis of the experts and the decision maker, that was regarded as the best means of achieving the aim of reducing congestion in the CCZ.
52. As a matter of law, there is no doubt that more than one aim can be pursued by a given measure and perhaps more importantly, a dual aim which appears to pull in two different directions can nonetheless be a legitimate aim: see for example European Commission v Hungary 6 November 2012 C-286/12, which concerned potentially unlawful age discrimination in a scheme requiring compulsory retirement of judges and lawyers, and where the CJEU held that a dual aim of establishing a balanced age structure for the young and older civil servants while at the same time seeking to provide a high quality justice service (staffed by experienced and by implication older civil servants) could constitute a legitimate aim. The mere fact that the aim here is to reduce congestion while maintaining the existing level of wheelchair accessible passenger vehicles does not make it an illegitimate aim. Instead, the extent to which there is an adverse impact on some disabled people is an issue to be addressed at the proportionality stage.
53. Having considered the material with care, I am satisfied that it cannot be said that maintaining the current level of wheelchair accessible vehicles is not an important and legitimate aim in this context. The importance of providing transport for wheelchair users who cannot transfer from their wheelchair to a seat, and for those who can, providing a choice as to how they travel (by remaining in the wheelchair or transferring to a seat) was recognised by Parliament when imposing the duty in s.165 of the Equality Act 2010 on all designated wheelchair accessible vehicles (in other words in London, all taxis and wheelchair accessible minicabs), "to carry the passenger while in the wheelchair", "not to make any additional charge for doing so", "to take such steps as are necessary to ensure that the passenger is carried in safety and reasonable comfort" and to "give the passenger such mobility assistance as is reasonably required". Only wheelchair accessible vehicles provide this choice and introducing a disincentive for wheelchair accessible passenger vehicles (in the form of liability to congestion charge) would have run the risk of reducing the number of such vehicles for a group that is particularly dependent on the availability of them. As a matter of logic, any reduction in wheelchair accessible transport in the CCZ would reduce the availability of vehicle transport to wheelchair users and make it harder to obtain such transport.
54. This is reinforced by ss.142(2)(a) and 142(4) of the 1999 Act which require the Mayor to detail in his transport strategy proposals for the provision of transport which is accessible to people with mobility problems; and consult with the Disabled Persons Transport Advisory Committee and with other people or bodies representing the interests of people with mobility problems.
55. As to the evidence about levels of wheelchair accessible vehicles within the CCZ, Ms Calderato gave evidence of an insufficiency of wheelchair accessible vehicles generally and had no reason to think the London CCZ is any different. In support of that conclusion, she identified calls and complaints (see in particular her second witness statement at [45]) from different disability groups (and others) for an increase in wheelchair accessible transport and for tackling the dearth of wheelchair accessible minicabs. This evidence was accepted by the judge.
56. Further, it is undoubtedly the case that the judge considered the position of disabled passengers generally, recognising expressly the impact of higher prices and lower numbers of minicabs willing to travel into the CCZ on this vulnerable group of passengers. He dealt with this aspect of the evidence as part of his assessment of whether the measure was a proportionate means of achieving the aim, rather than whether it was appropriate and therefore legitimate. I agree with his approach. The fact that a measure achieves benefits for one group of disabled people but disadvantages another group of disabled people does not by itself render the measure inappropriate or unsuitable to achieve the aim (although I can see that there might come a point where a measure is so incoherent or inconsistent in relation to its impact on different groups with the same protected characteristic that it might be said to be an irrational way of seeking to do so). That is however not this case. In my judgment the impacts, both positive and negative, for different groups of disabled people and others, fell properly to be considered at the proportionality stage of the analysis when considering whether the measure is a proportionate means of achieving the aims identified as legitimate. That is entirely consistent with the approach of the Advocates General in Dansk and VL.
57. For all these reasons and notwithstanding my initial concerns, I am satisfied that Lewis J was entitled to conclude that the measure was an appropriate means of achieving the aim of reducing traffic and congestion in the CCZ without reducing the number of wheelchair accessible passenger vehicles in the CCZ and made no error in doing so.

Ground 3: No Other Less Intrusive Means

58. Lewis J concluded that the Mayor had established that there were no other less intrusive measures that could realistically achieve the same aim. He reasoned as follows:

"76. … Various suggestions of alternative measures had been raised by the claimants in their claim form but were not pursued in their written or oral submissions. I am satisfied by the evidence of Ms Calderato that those measures were not available measures. There was a suggestion that the congestion charge could be increased for all those presently liable to the charge (but private hire vehicles remaining exempt) and that might result in a reduction in traffic of 1% overall. First, that would not, in fact, address the increase in the number of private hire vehicles in the CCZ, nor the fact that a large number circulate without passengers. The aim is to reduce the number of such vehicles in the CCZ, by making changes which will either encourage fewer journeys into the CCZ or which will result in more efficient and better utilisation of private hire vehicles within the CCZ (fewer vehicles undertaking the journeys required, and fewer vehicles present in the CCZ without passengers). Measures aimed at producing a 1% reduction in traffic overall would not, of themselves, address the particular problems of the increase in the number of private hire vehicles in the CCZ. Furthermore, the evidence of Ms Calderato is that even a large increase in the congestion charge for those currently liable to pay it is unlikely to yield substantial reductions in the number of vehicles in the CCZ. Ms Calderato points out that an increase from £5 to £8 (more than 50%) in 2005 did not lead to any discernible reduction in traffic."

59. Mr Collins contends, correctly, that it was for the Mayor to show that the measure adopted was a proportionate means of achieving a legitimate aim and not for the appellant to establish an obviously better alternative: see Hockenjos v Secretary of State for Social Security [2004] EWCA Civ 1749, [2005] IRLR 471 at [47]. He submits that Lewis J erred by limiting his evaluation of alternative measures to those suggested by the appellant, and accordingly made no proper finding. Furthermore, the judge wrongly elided the aim of reducing traffic and congestion more generally with a need to reduce the number of minicabs, leading him to focus on the wrong question and vitiating his conclusion that there were no less intrusive alternative means that could have been adopted.
60. I have already dealt with the question whether the first aim was misstated by the Mayor. I do not think it was. Nor was it mischaracterised by the judge. Rather, Lewis J recognised, as do I, that the expert analysis and evidence showed that a measure aimed at producing a 1% reduction in traffic overall would not by itself, address the particular congestion problem caused by the dramatic and unforeseen increase in the number of exempt minicabs in the CCZ at a time when the number of private cars in the CCZ was in fact falling, and the number of taxis was relatively stable. In other words, if minicabs were not deterred from entering the CCZ in the particular circumstances, the overall traffic reduction or congestion reduction benefits were not likely to be achieved. In those circumstances it is unsurprising that Lewis J referred interchangeably to reduction of traffic and reduction in the number of minicabs. That was not because he wrongly elided the two; but because without a reduction in minicabs entering the CCZ, the evidence was that no reduction in traffic was likely to be achieved.
61. There is no dispute as to the proper approach to the question of less restrictive but equally effective alternative measures. It was considered by this court in R (The Friends of Antique Cultural Treasures Ltd) v Secretary of State for the Department of Environment, Food and Rural Affairs [2020] EWCA Civ 649 where the main points derived from case law were summarised in the following way:

"80. … This is an area where the respondent's margin of appreciation or discretion is relevant. The main points arising from case law can be summarised as follows:

(i) The decision maker has a margin of appreciation or discretion which is highly fact and context specific: Lumsdon paragraphs [64] and [65]. The evaluation will take account of all relevant circumstances including the conditions prevailing in the relevant market, the circumstances leading up to adoption of the challenged measure, and the reasons given why less restrictive measures were rejected.

(ii) A measure will be disproportionate if "it is clear that the desired level of protection could be attained equally well by measures which were less restrictive": Lumsdon paragraph [66]; EU Lotto paragraph [104].

(iii) The burden of proof lies with the decision maker. It is not however to be applied mechanically. There is no duty on the decision maker to prove positively that no other measure could be as effective: Lumsdon paragraph [63]; Scotch Whisky paragraph [55]; BAT (ibid) paragraph [659].

(iv) The decision maker is not required "… to consider every possible alternative, including those that were never suggested by consultees": TfL paragraph [37]; EU Lotto paragraph [104].

(v) The mere assertion that some other measure is equivalent and less intrusive is not sufficient: BAT (ibid) at paragraph [662]; and equally the fact that some other measure can be envisaged is not enough: BAT (ibid) paragraphs [660] – [662].

(vi) It is relevant that a measure is "general, simple, easily understood and readily managed and supervised": BAT paragraph [661]."

62. Here, Lewis J noted that a number of suggested alternative measures were identified by the appellant in the claim form: charging operators, cap and control licensing, introducing wage protection and/or the provision of rest areas within the CCZ. Those were all addressed in the evidence of Ms Calderato, which explained why none of these suggestions was legally or practically feasible or liable to secure the same benefits as the measure. None of these suggestions was pursued in argument below. Instead the appellant's skeleton argument raised for the first time the question whether the congestion charge should have been increased for all those presently liable to the charge (but with minicabs remaining exempt). It was argued that this might result in a reduction in traffic of 1% overall without the adverse impacts that removal of the PHV exemption would entail.
63. This suggested alternative was addressed in Ms Calderato's second witness statement. She disagreed with the suggestion that a 1% increase in the charge overall was likely to achieve the traffic/congestion benefits that the measure was likely to achieve. She explained that this was because:

"(1) … the number of PHVs in the CCZ has grown significantly in a sustained manner, whereas other exempt vehicles have not.… TfL's 2017 surveys suggested that 26% of PHVs circulate without passengers in the CCZ. Prior to the [measure] being implemented, PHVs were all exempt from the congestion charge. There was therefore no disincentive to them from entering the CCZ, and they were often in the CCZ while not carrying passengers. In consequence, it is possible in principle to reduce traffic in the CCZ by imposing the congestion charge on PHVs, and CEPA has estimated in detail by how far.

(2) … ITPL observed in their 2017 report that reductions in personal motor vehicles in London appeared to have been offset by increases in PHVs and other vehicles. More generally, ITPL explained that common experience in congestion management around the world is that the creation of new road capacity tends to induce further demand.…

As ITPL summarised… policies that cause "increased road capacity" will "generate traffic". ITPL further explained that interventions of this kind, which lead to an overall increase in traffic, would themselves often lead to future interventions being needed.…

ITPL said that "best practice in scheme design would point to there being as few exemptions as possible" in the congestion charging scheme… TfL agrees with ITPL's analysis.

(3) … the number of PHVs in the CCZ has steadily increased… and there is a large number of PHVs in London that have has also been increasing until recently… Applying the principles just explained, I consider it very unlikely that increasing the congestion charge on other vehicles already subject to it but maintaining the discounts and exemptions in place before 8 April (in particular the exemption for PHVs) would, overall, achieve the same traffic/congestion benefit as removing the exemption from a large class of exempt vehicles, namely, non-wheelchair accessible PHVs. ITPL's analysis and the circumstances I have mentioned indicate that any road capacity freed up by increasing the congestion charge for other vehicles … would very likely be offset by additional PHVs entering the CCZ."

64. Lewis J accepted Ms Calderato's evidence. He accepted that consideration was given to the alternatives suggested by the appellant in the claim form, but none of these was available.
65. As for the suggested general increase in the congestion charge (while maintaining the exemption for minicabs), he concluded on the basis of her evidence that "even a large increase in the congestion charge for those currently liable to pay it is unlikely to yield substantial reductions in the number of vehicles in the CCZ." That was a conclusion he was entitled to come to in light of the evidence, and for the reasons already given, the judge was not focused on the wrong aim but recognised that a reduction in the number of minicabs was the means of achieving the overall aim of reducing traffic congestion.
66. In reaching that conclusion I have considered the evidence of a recent, temporary increase in the congestion charge for all vehicles liable to pay it, in response to the Covid 19 pandemic. This was part of a package of temporary measures (that included extending the operating hours for the congestion charging scheme each day to 10pm and to cover Saturdays and Sundays) to incentivise vehicles not to enter the CCZ following the lifting of lockdown and in circumstances where there was strong evidence of a reluctance among the public to use public transport. In the circumstances, I do not consider that the adoption of this package of temporary measures casts any doubt on the decision reached by Lewis J on the material then before him. Moreover, there was no other clear or obvious alternative that would achieve the desired reduction in a less intrusive way and nobody suggests that the Mayor was bound to consider every possible alternative, including those never identified or suggested. As Mr Collins accepted, there are limits to the burden on a decision maker in this regard and no duty to prove positively that no other measure could be as effective.
67. Accordingly, applying the principles established by the case law (as set out above) I can see no error in the approach or analysis of Lewis J in relation to this issue.

Ground 4, 6 and 7: Proportionality

68. I take these three grounds together because they all concern the legality of the judge's approach to the question of proportionality, in other words the question whether the measure is a proportionate means of achieving the legitimate aim of reducing traffic congestion within the CCZ whilst maintaining the current levels of wheelchair accessible vehicles for disabled passengers, and his conclusion that proportionality had been established.
69. The proper approach to the question whether a measure is a proportionate means of achieving a legitimate aim is set out above.
70. Before this court there was some disagreement between the parties as to the intensity of the scrutiny required, and as to the extent of the margin of discretion (or appreciation) to be accorded to the decision maker in this context. Ms Demetriou submitted that the application of the proportionality test itself imported a stringent standard of scrutiny and no more was required. Furthermore she submitted that the Mayor enjoys a margin of appreciation as to both "the level of protection of the public interest in question" and his "selection of an appropriate means by which that protection can be provided": see R (Lumsdon) v Legal Services Board [2015] UKSC 41, [2016] AC 697 at [66]. So far as the balance between competing interests under the Equality Act 2010 is concerned, she submits that there is a wide margin of discretion to be accorded to the Mayor for four reasons. First, he is democratically elected and accountable. Secondly, he has undertaken to keep the congestion charging scheme under review. Thirdly, the decision-making process involved a predictive and evaluative judgment for which the Mayor and Transport for London repeatedly sought and relied on expert assistance from CEPA. It was objectively reasonable to conclude on the basis of the CEPA analysis and the other material available to the Mayor, that the measure would have the effects predicted. The margin of discretion extends to the judgment in choosing the means by which the object is achieved. Finally, she relies on the fact that a lack of regulation would be harmful to road users.
71. In my judgment and in agreement with Mr Collins a rigorous standard is required in scrutinising the justification advanced by the respondent in this case for the reasons given by Mummery LJ in R (Elias) v Secretary of State for Defence [2006] EWCA Civ 1293, [2006] 1WLR 3213. At [161] Mummery LJ held that:

"a stringent standard of scrutiny of the claim to justification is appropriate because the discrimination, though indirect in form, is so closely related in substance to the direct form of discrimination on grounds of national origins, which can never be justified."

So too here given the statistical imbalance in the ethnic composition of the two groups (taxi drivers and minicab drivers). This is stark: the measure has no impact on the predominantly white taxi driver group and an adverse impact on the group of minicab drivers from predominantly black and minority ethnic backgrounds.

72. That does not mean there is no scope for the margin of discretion to operate. As the appellant accepts, in general, public bodies have a wide margin of discretion in determining whether decisions in the field of social or economic policies are proportionate to a legitimate aim. In relation to both the identification of a potentially legitimate aim and the assessment of proportionality, the court must make a rigorous objective assessment for itself, but in making that assessment, it must accord an appropriate margin of discretion to the decision maker having regard to the circumstances of the particular case: see McCloud v Lord Chancellor [2018] EWCA Civ 2844, [2020] 1 All ER 304 at [143 to 145].
73. Turning to the grounds of challenge, the first focusses on paragraph 77 of the judgment, where Lewis J indicated that he was moving to the question of proportionality and said:

"The issue here is whether the defendant can demonstrate that the impacts on BAME and female drivers and disabled passengers are justified as a proportionate means of achieving the aim. It is the impact on those groups that has to be justified not the measure. The starting point is to consider the likely impacts on those affected".

74. Mr Collins criticises the underlined sentence. He relies on Baroness Hale's statement in R (UNISON v Lord Chancellor [2017] UKSC 51, [2017] 3 WLR 409 at [126], that in relation to indirect discrimination, "it is the PCP itself which requires to be justified, rather than its discriminatory effect" and submits that Lewis J was wrong to say it is the impact on affected groups that must be justified not the measure. It is the measure that must be justified. He submits that this error led to the adoption of the wrong framework for what followed, and meant that the incorrect legal test was applied and there was a failure to evaluate the PCP itself.
75. Furthermore, Mr Collins submits that the judge did not weigh the measure against the benefits to be achieved by it. He set out the impact of the measure but, Mr Collins submits, the judge never conducted the necessary critical evaluation by balancing its moderate benefits against the serious impacts to many of the most vulnerable, low income people within the capital. Instead, he simply listed the factors and then stated his satisfaction that the respondent had demonstrated that removal of the exemption from PHV drivers was a proportionate means of achieving a legitimate aim.
76. Mr Collins also submits that when conducting the proportionality exercise, Lewis J afforded the respondent too wide a discretion and fell into error as a consequence. He accepts that discretion in relation to both aims and means is afforded to decision-makers but contends that Lewis J went too far here in allowing the public nature of the respondent to determine the question of proportionality. Even where there is a question of social policy involved, that does not remove the requirement on the court to conduct the necessary careful balancing exercise. That is particularly so in circumstances where the discrimination involved is on a suspect ground so that greater scrutiny is required. Here, the context is drivers and passengers with protected characteristics of race, sex and disability, and the interference goes to aspects of their personal integrity, affecting their status as citizens in London. The magnitude of the impact was not properly weighed in the balance against the minimal benefits to be achieved.
77. I start with the asserted inconsistency between the approach identified by Lewis J at paragraph 77 and the statement made by Baroness Hale in UNISON. In my judgment there is no inconsistency: they were referring to different, albeit to some extent overlapping, stages of the relevant enquiry. As s.19(2)(d) of the Equality Act 2010 makes clear, the putative discriminator must show that the PCP is a proportionate means of achieving a legitimate aim in order to justify what would otherwise be unlawful indirect discrimination. That involves determining first whether the measure (or PCP) is directed at achieving a legitimate aim; in other words, an aim that corresponds to a real need, is appropriate to achieving the objective in question and reasonably necessary to achieve the aim. Secondly, the measure adopted must be a proportionate means of achieving that aim. Although the terms proportionate and justified are often used interchangeably (as indeed appears from the guidance derived from the case law in Lockwood as set out at paragraph 36 above), they are different. Justification involves two stages, the second of which involves the application of the proportionality principle. Once that is appreciated, it seems to me to be clear that when Baroness Hale referred to the requirement that the PCP be justified, she was referring to the overall approach in s.19(2)(d); whereas, Lewis J had dealt with the first aspect of justification, namely the legitimacy of the aim in earlier paragraphs of his judgment (paragraphs 74, 75 and 76 in particular) and was turning (at paragraph 77) to address the second stage, namely the proportionality stage of the justification enquiry. Further, as is well settled, the test of proportionality is in essence, a balancing exercise between the discriminatory effect of the measure (or PCP) on the disadvantaged group, and the needs of (or benefit to be achieved by) the putative discriminator on the other. As Mummery LJ put it in Elias at [151] "it is necessary to weigh the need against the seriousness of the detriment to the disadvantaged group." In other words, the judge had to evaluate the extent of the adverse impact of the measure adopted by the respondent on individuals with the protected characteristics of race, sex and disability, weigh that against the benefits and importance of the measure and determine where the balance lay. For these reasons the approach set out by Lewis J at paragraph 77 reflects no error of law on his part.
78. As to the proportionality analysis itself, I have come to the conclusion, after close analysis of all the material in this case, and notwithstanding the forceful arguments advanced by Mr Collins to the contrary, that the judge made no error in conducting the balancing exercise required by law. It seems to me that the stark statistical imbalance in this case led both the Mayor (and Transport for London) and Lewis J to confront this concerning feature of the measure adopted, and to scrutinise with particular care the nature and significance of the impact on the disadvantaged groups when balancing that against the aims and objects to be achieved.
79. At paragraphs 79 to 84 Lewis J conducted a careful analysis of the likely impacts on the disadvantaged groups of ethnic minority (and other) minicab drivers and on passengers, and in summary, he found that:

a) the Mayor (and Transport for London) plainly recognised the adverse impact on minicab drivers, 71% of whom live in the most deprived areas of London, with mean annual earnings less than £23,000 per annum net; and 2% of whom are women. The judge considered the impact on these drivers together. He identified the maximum financial consequence that could arise as a result of removal of the exemption from the congestion charge (assuming the driver paid the charge, and used Auto Pay) as £52.50 a week, or approximately £230 a month, in congestion charges (in other words, a loss of about 10% of net income).

b) However, this impact was mitigated by three principal matters.

c) First, the evidence showed that two thirds of London region minicabs never enter the small (but significant) central area of London that is the CCZ. They work, and undertake journeys, in the significantly bigger area outside the CCZ, constituting Greater London and so would not see a substantial increase in operating costs through paying the congestion charge. The judge did not ignore the fact that one third of drivers would remain, in principle, affected and I accept, as Mr Collins submitted, that this is not an insignificant group. However, it is significantly smaller than the London region minicab drivers as a whole and it was legitimate to have regard to this factor.

d) Secondly, it was reasonable to expect (as the respondent did) that operators would make changes enabling drivers to recover some of the cost of the congestion charge thereby reducing its impact (and in fact, Uber, the largest PHV operator, had done so, and now levies a charge of £1 on all journeys passing through the CCZ, whether made during charging hours or not, and passes the levy to drivers). The judge recognised that the extent to which such a levy reduces the impact on a driver depends on the number of times the driver carries a passenger within the CCZ. There was also evidence that another large operator (Addison Lee) with a different operating model, pays the charge as registered keeper of their vehicles but has introduced an increased rental fee of £15 a week charged to its drivers for leasing the vehicle. The reduced impact on those drivers will be about £15 a week (rather than £52.50).

e) Thirdly, there was evidence of some operators modifying their operating practices so as to enable drivers to avoid going into the CCZ if they wish and thereby to avoid having to pay the congestion charge. For example, there was evidence that ViaVan allows drivers to indicate willingness to drive into the CCZ, and those choosing not to are not affected by the congestion charge. There was evidence that the provision by Uber of postcodes for the destination enables drivers to decide if they wish to accept the fare and, if the postcode is known to be within the CCZ, they can decide not to accept it. Again, and accepting that it may take time for postcodes to be recognised by minicab drivers as within or outside the CCZ, this was a legitimate factor to consider. The judge also recognised that some drivers may continue to go into the CCZ because, some at least, will calculate that they can do enough journeys in the CCZ with the levy to cover the cost, or a sufficient part of the cost, of the congestion charge to make it worthwhile.

f) For those who remain likely to be economically adversely affected to some degree by the withdrawal of the exemption from the congestion charge, the judge accepted that their income would reduce or they would have to work longer hours to meet their basic costs including payment of the congestion charge if they enter the CCZ. He set out the particular impact this was having on the individual claimants, Mr Ali and Mrs Minshull, and their respective families.

g) The position of disabled passengers was separately addressed both by the Mayor and the judge. The evidence anticipated an increase in the cost for them of journeys into the CCZ (where, for example, all or part of the congestion charge is passed on to customers) or there might be a reduction in the availability of PHVs willing to travel into the CCZ. The likely increase was thought to be £1 to £2 a journey. The Request recognised that an increase of even £1 or £2 would not be an insubstantial amount for some disabled people. The contemporaneous material suggested some of the impact could be mitigated by schemes offering subsidised taxi and PHV journeys. However, for disabled people whose disabilities do not include mobility issues (or disability issues relating to transport), they may be forced to use less convenient but cheaper public transport. There was a likely adverse impact on disabled people in consequence.

80. In my judgment, there was a careful examination by Lewis J of the impact on each of the disadvantaged groups and a critical evaluation of it and how it might be ameliorated, and where that was unlikely, the extent of the remaining impact. I can detect no error in the judge's approach to this assessment.
81. Before coming to his assessment of proportionality, and as part of it, Lewis J had regard to the fact that the Mayor is elected and accountable for the decisions taken in relation to the management of road user charging systems within London. He was informed of the potential effect of the removal of the exemption on drivers from minority ethnic backgrounds and part-time drivers who were likely to be women. He considered that the importance of the aim – the reduction of traffic and congestion in Central London – did justify the potential impact on the drivers concerned. His approach is challenged as wrong in law because it is said that Lewis J allowed the public nature of the respondent to determine the question of proportionality rather than doing so himself. Mr Collins contends that at paragraph 85 the judge, in effect, deferred to the Mayor and found it sufficient that the Mayor had turned his mind to the matters in issue. That was to misunderstand his role and to afford too wide a discretion to the respondent.
82. I do not accept these arguments. Although this is a case in which an increased level of scrutiny is called for, that does not mean no margin of discretion is to be accorded to the decision maker. To the contrary, in my judgment the Mayor had some margin of discretion both as to the aims and as to means adopted to achieve them. It was for Lewis J to determine the breadth of the margin in each case because of the fact sensitive nature of this issue. That is precisely what he did. He identified the features of the case that led him to conclude that some weight was to be accorded to the democratically accountable decision maker who determined the appropriate balance to be struck between the competing interests in this case. Those were both relevant and legitimate factors that Lewis J was entitled to consider, although I echo the caution expressed by Singh LJ at paragraph 94 below as to the importance of the Mayor's democratic accountability in the present stark context. The judge did not accept the Mayor's views uncritically. Instead, he conducted his own proportionality assessment that had regard, as a relevant but not a determinative factor, to the margin of discretion to be accorded to the Mayor.
83. Mr Collins criticises the fact that at paragraph 86, when reaching his conclusion that the Mayor had discharged the burden of establishing that removal of the exemption from the congestion charge for minicabs (save for those that are wheelchair accessible and being used as such) was a proportionate means of achieving a legitimate aim notwithstanding the impact on the disadvantaged groups identified, Lewis J did not weigh the minimal benefits of the aim against the significant disadvantages caused by the means used to achieve it. I accept that this balancing exercise was not expressly stated. However, it is clear from the analysis as a whole, that this is precisely what he did. In the earlier parts of his judgment, the judge accepted the scale of the growing congestion problem for London and that there were good reasons to focus on the increasing numbers of minicabs in London. He grappled with the 1% traffic reduction forecast, and accepted its reliability. He found that the measure corresponded to a real need and would achieve a not insignificant benefit and that there were no other less intrusive measures which could realistically achieve the same aim. It is entirely unrealistic to suggest that he overlooked that side of the balance when he came to assess proportionality, having critically assessed the negative impact of the measure in the way that I have summarised above.
84. Having scrutinised the material available to the judge and his reasoning with particular care, it seems to me that Lewis J reached a justified conclusion that the not insignificant benefits forecast (for traffic reduction, congestion and air pollution) for Londoners outweighed the likely adverse impact on a relatively small proportion of minority ethnic and female drivers, and disabled passengers, so that removal of the exemption from the congestion charge for minicabs (other than those designated wheelchair-accessible vehicles) is a proportionate means of achieving the legitimate aim in all the circumstances.
85. For all the reasons given above, I am satisfied that no error has been shown in Lewis J's approach, reasoning or conclusions on the proportionality aspect of the appeal. In reaching my conclusion I have considered all points made on behalf of the appellant on this aspect of the appeal, but in so far as additional points were made, I need not further lengthen this judgment by addressing them here. It is sufficient to state that I have not been persuaded that any flaw in Lewis J's judgment has been shown.

Ground 5: The comparator pool

86. I can deal very shortly with this ground. Mr Collins contends that Lewis J approached s.19(2)(b) Equality Act 2010 incorrectly when identifying the pool of comparators for the purposes of the unlawful indirect discrimination claim. The appellant argues that Lewis J wrongly considered the position of black and minority ethnic minicab drivers (and female drivers) relative to all other minicab drivers, whereas he should have considered their position relative to all minicab and taxi drivers. The failure to identify the correct comparator pool infected the rest of the judge's analysis.
87. I do not accept that Lewis J made the error contended for here. It seems to me to be clear that for the purposes of the judgment he assumed that s.19(2)(a), (b) and (c) were all satisfied. In particular, at paragraph 66, Lewis J said in terms, "For present purposes, I assume that the amendments to the Scheme … fall within section 19(2)(a)(b) and (c) so far as BAME persons, women and disabled passengers are concerned'. He proceeded on that basis without ruling determinatively on the appropriate comparator pool.
88. For these reasons, this ground also fails.
89. In these circumstances, and if my Lords agree, the appeal fails and must be dismissed. In those circumstances it is unnecessary to deal with the point raised by the Respondent's Notice. The point is not without difficulty and in circumstances where it is unnecessary to deal with it, I prefer to express no view on it.

Singh LJ:

90. I agree that this appeal should be dismissed for the reasons given by Simler LJ. I would like to add a few words of my own in view of the troubling nature of this case. There are three topics on which I would like to add some observations.
91. The first topic is the question of the aims of the measure under challenge. I accept that a measure can have more than one legitimate aim. The difficulty in the present case is that the second aim relied upon does not, on objective analysis, appear to be an aim at all. If the Respondent did not have the first aim, that is of reducing congestion, the second aim would not exist independently of it, that is to maintain the current level of wheelchair-accessible transport vehicles in London. As a matter of objective analysis, it is arguable that the second aim is in truth a design feature of the means by which the first aim was to be achieved. In that sense it was a non-negotiable way of achieving the first aim. It could be argued that, when assessing whether there is objective justification for what is on the face of it indirect discrimination, all aspects of the design features employed need to be subjected to appropriate scrutiny. Otherwise, there is a risk that the inclusion of a design feature as one of the aims will preclude proper scrutiny of whether that design feature is a justified one. Ultimately, however, I am persuaded that, even if the second aim is properly to be regarded as part of the means rather than an aim in itself, there were legitimate reasons for distinguishing between minicabs and taxis. This is essentially for the reasons set out by Simler LJ at paragraph 50 above.
92. The second topic is the issue of whether there were less restrictive means available to the Respondent. I find it surprising that no consideration appears to have been given early on to the question whether the congestion charge should have been increased for all those presently liable to the charge (but with minicabs and taxis remaining exempt). It seems to me that that was an obvious possible alternative solution to the perceived problems that the measure was intended to resolve. Nevertheless, once this suggested alternative was raised by the Appellant, in its skeleton argument in the High Court, it was addressed by way of evidence in Ms Calderato's second witness statement. I agree with Simler LJ that Lewis J was entitled to accept her evidence that even a large increase in the congestion charge generally was unlikely to yield substantial reductions in the number of vehicles in the CCZ. This Court, sitting as an appellate court, cannot say that the judge was wrong to accept that evidence.
93. The third topic is the question of the appropriate margin of discretion or judgement to be afforded to the Respondent. The Respondent relies in particular on what has been said in cases such as R (Lumsdon) v Legal Services Board [2015] UKSC 41; [2016] AC 697. I agree with Simler LJ that we should reject that approach and adopt the "stringent scrutiny" approach which was set out by Mummery LJ in R (Elias) v Secretary of State for Defence [2006] EWCA Civ 1293; [2006] 1WLR 3213, at [161]. This is particularly because the impact on drivers from black, Asian and other minority ethnic ("BAME") communities as compared with white drivers is striking. As Simler LJ has mentioned, 94% of minicab drivers are from BAME communities, whereas only 12% of taxi drivers are. What is in issue in the present case is alleged discrimination, in particular on racial grounds. Whatever may be the position generally when reviewing the acts of public authorities in the context of social and economic policy, in my view, a more stringent scrutiny is required when the alleged ground of discrimination is race.
94. I am also unimpressed in the present context with the reliance which has been placed on the fact that the Respondent is democratically elected. The democratic nature of our society is of vital importance and must be respected by the courts. But the premise of equality law is that every person is entitled to be treated in an equal way irrespective of whether they are part of the majority or in a minority. In a democracy there is a danger that the majority (perhaps unconsciously) will override the interests of the minority because the price of a measure will be paid not generally by the community but only, or substantially, by a minority.
95. If a measure discriminates directly on grounds of race, it can never be justified under the Equality Act. If what is alleged is indirect discrimination, in principle it is capable of justification but the court should be alert to the need to prevent the danger of unconscious bias by subjecting the justification which is put forward to appropriate scrutiny.
96. It is also important to recall in this context that democracy is not the same thing as majority rule. As Baroness Hale explained in a human rights case, Ghaidan v Godin-Mendoza [2004] UKHL 30; [2004] 2 AC 557, at [132]:

"Democracy values everyone equally even if the majority does not."

97. Nevertheless, for the reasons set out by Simler LJ, I am persuaded that, even applying a rigorous standard of scrutiny in this case, the Respondent was entitled to adopt the measure under challenge.

Sir Geoffrey Vos, Chancellor of the High Court:

98. I agree with both judgments.
99. I too found the case troubling. At first sight, the measure adopted by the Mayor looked as if it might have been targeted at (mostly BAME) minicab drivers who are deprived of the exemption, leaving (mostly white) taxi drivers exempt from the congestion charge. It seemed to us, as we have now held, that stringent scrutiny would be required to justify such a state of affairs.
100. On careful analysis, however, I have been persuaded that appearances were indeed deceptive. First, there is no real doubt that the Mayor's overwhelming and legitimate aim was to reduce traffic and congestion in the zone. Secondly, it became clear that the only effective way of doing so would be to address the exponential rise both in the absolute number of minicab drivers operating in London and in those entering the zone. Fewer private cars and taxis were doing so, but minicabs continued to increase. Thirdly, there was no purpose in interfering with the exemption accorded to taxis, because taxis provide unique advantages to Londoners. They allow wheelchair bound passengers to travel and are required to carry passengers under a raft of regulations that do not affect minicabs.
101. In short, the measure and its discriminatory impact on BAME minicab drivers was, in my judgment, justified by the legitimate aim of reducing traffic, congestion and pollution. The (seemingly slim) 1% traffic reduction forecast turned out to be a little misleading, since, if nothing were done to stem the growth of minicabs in the zone, overall future traffic increases were inevitable.