37
1 | Page Policy D4: Development in urban areas and Inset Villages I OBJECT to this policy. Submission Summary There are effectively two separate parts to this policy, both are flawed, and in the case of insetting, fatally flawed. In the case of urban development there are numerous flaws, including: The plan envisages a large expansion of the traditional “bricks and mortar” retail sector, but this is a sector in long term decline, and so is not sustainable even in the medium term. The plan envisages that 40% of homes built will be affordable, but these will not provide appropriate accommodation for low paid workers, for example, those on the minimum wage, who cannot afford to buy “affordable” homes, nor can they pay so-called “affordable” rent. There is a great need for an expansion in social housing, especially in urban areas, where there is a greater concentration of low paid workers such as those employed in the retail, warehousing and distribution sectors. The number of homes planned in the urban area is too low – there is a need for regeneration in some areas of Guildford, but this is ignored and instead the plan concentrates on building homes on greenfield sites, which does not meet the needs for housing within the town, especially within the centre. This is not a sustainable policy as building outside the town has a disproportionate effect on road traffic. The analysis in Annexe 1 shows this to be the case when alternative methods of travel to work are analysed for two separate areas, one in the town centre and the second in a rural village. In the case of insetting, the policy is fatally flawed, for three separate reasons. Firstly, it is based on an erroneous assumption. GBC justify the policy (in para 4.5.47) by suggesting, without any supporting data, that development in villages washed over by the green belt has been at a lower rate than within Guildford town. This was not the case, and statistics from the ONS censuses of 2001 and 2011 show that this assumption is incorrect. These statistics show that the growth rate in housing was higher in Effingham, a village washed over by the green belt, than in Merrow, which is not washed over by the green belt. In fact, the rate of increase in housing stock in Effingham was faster than the average for the borough. Secondly, insetting has been based on the Green Belt and Countryside Study, but the analysis in this document was flawed to the extent that it is valueless. It has serious flaws in its methodology, includes subjective judgements, and thus its conclusions are flawed - see Annexe 2 for detailed analysis These flaws are so serious that this study should be removed from the evidence base that supports the Local Plan. Finally, although a flawed justification was provided for the insetting of villages, this does not apply to all the areas selected for insetting, many of which are not “villages”, such as Pirbright Barracks. These sites are listed in the text below. No attempt has been made to justify the insetting of these sites, and in the absence of any justification they should remain in the Green Belt.

Policy D4: Development in urban areas and Inset Villages I

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Policy D4: Development in urban areas and Inset Villages I

1 | P a g e

Policy D4: Development in urban areas and Inset Villages

I OBJECT to this policy.

Submission Summary

There are effectively two separate parts to this policy, both are flawed, and in the case of insetting,

fatally flawed. In the case of urban development there are numerous flaws, including:

The plan envisages a large expansion of the traditional “bricks and mortar” retail sector, but

this is a sector in long term decline, and so is not sustainable even in the medium term.

The plan envisages that 40% of homes built will be affordable, but these will not provide

appropriate accommodation for low paid workers, for example, those on the minimum

wage, who cannot afford to buy “affordable” homes, nor can they pay so-called “affordable”

rent. There is a great need for an expansion in social housing, especially in urban areas,

where there is a greater concentration of low paid workers such as those employed in the

retail, warehousing and distribution sectors.

The number of homes planned in the urban area is too low – there is a need for

regeneration in some areas of Guildford, but this is ignored and instead the plan

concentrates on building homes on greenfield sites, which does not meet the needs for

housing within the town, especially within the centre. This is not a sustainable policy as

building outside the town has a disproportionate effect on road traffic. The analysis in

Annexe 1 shows this to be the case when alternative methods of travel to work are analysed

for two separate areas, one in the town centre and the second in a rural village.

In the case of insetting, the policy is fatally flawed, for three separate reasons.

Firstly, it is based on an erroneous assumption. GBC justify the policy (in para 4.5.47) by

suggesting, without any supporting data, that development in villages washed over by

the green belt has been at a lower rate than within Guildford town. This was not the

case, and statistics from the ONS censuses of 2001 and 2011 show that this assumption

is incorrect. These statistics show that the growth rate in housing was higher in

Effingham, a village washed over by the green belt, than in Merrow, which is not washed

over by the green belt. In fact, the rate of increase in housing stock in Effingham was

faster than the average for the borough.

Secondly, insetting has been based on the Green Belt and Countryside Study, but the

analysis in this document was flawed to the extent that it is valueless. It has serious

flaws in its methodology, includes subjective judgements, and thus its conclusions are

flawed - see Annexe 2 for detailed analysis These flaws are so serious that this study

should be removed from the evidence base that supports the Local Plan.

Finally, although a flawed justification was provided for the insetting of villages, this

does not apply to all the areas selected for insetting, many of which are not “villages”,

such as Pirbright Barracks. These sites are listed in the text below. No attempt has been

made to justify the insetting of these sites, and in the absence of any justification they

should remain in the Green Belt.

Page 2: Policy D4: Development in urban areas and Inset Villages I

2 | P a g e

Urban development

There is a fundamental flaw in the draft Local Plan, in that there are not enough dwellings planned in

urban areas, in particular, Guildford town centre. This is where the demand for affordable homes is

most acute, a fact which is recognised in much of the draft Local Plan. Yet the plan then envisages

that most dwellings will be built outside the urban area, so there is no continuity in the plan – if the

demand for affordable homes is most acute within the town, then is where these dwellings should

be built. There are sites available, for example, various sites in Woodbridge Meadows, Walnut Tree

Close (including the area around the station) and North Street. Yet the focus in these areas is to

provide a very limited number of dwellings, with the emphasis on additional retail space and more

commercial premises. Given the profound changes in traditional retail, this is not a sustainable

policy. There have been many examples of bankruptcies amongst traditional retailers, the most

recent examples being BHS and Austin Reed, and previously Woolworths, Comet, etc. The British

Retail Consortium forecast that by 2025 up to one third of those now employed in traditional retail

could lose their jobs, due to a combination of factors1. These included the continuation of the move

to shopping via the internet and higher costs, from business rates and the introduction of the

“living” wage. Recent figures show that online sales are still growing quickly, with more than 13% of

total retail sales now made over the internet. Nor does an expansion in traditional retail chime with

the objective of attracting highly paid jobs to the area – the majority of jobs in retail are low paid,

and zero hours contracts are common. So any expansion in traditional bricks and mortar retail

would create a greater demand for affordable and social housing – exactly the reverse of what is

needed.

Even the major grocery retail chains have been affected by changes in retail habits – Tesco,

Sainsburys, Morrisons, etc are closing some stores, with no major large stores planned. Banks and

other financial institutions are still closing some branches, so changes in town centres still have

some way to go – but none of these changes are recognised in the draft Local Plan, which envisages

no limit to traditional retail.

There is another reason why the retail sector is unlikely to exhibit strong long term growth, and this

is consumer debt, which has reached a level that the Bank of England has said could contribute to

economic stability. This is likely to have a greater destabilising effect in the event of an economic

downturn – and there will certainly be at least two of these during the life of the Local Plan.

Consumer spending simply cannot be maintained indefinitely at its current rate, and to plan for a

large increase is irresponsible.

Building the majority of homes outside urban areas is contrary to the advice of planners such as Lord

Rogers, who has detailed the many advantages of increasing housing density in urban areas. The

housing policy in this plan will result in a greater number of commuter journeys, creating more

congestion, with an increase in air pollution across the borough. This is examined in detail in Annexe

1.

So there has to be a reappraisal of urban development plans, with more homes provided in urban

areas, in particular in the centre of Guildford. As the council own many sites in the centre, these

1 British Retail Consortium “Retail 2020: Fewer But Better Jobs” February 2016

http://www.brc.org.uk/downloads/Retail_2020_(final).pdf

Page 3: Policy D4: Development in urban areas and Inset Villages I

3 | P a g e

could be made available for development within the first 5 years of the plan, with mixed housing on

these sites, including new social housing, which is sorely needed. Affordable housing is simply not

affordable to anyone on the minimum wage, there is a need for a considerable expansion of social

housing, a need that is not adequately recognised in the draft Local Plan.

In the detail of Policy D4, it is suggested that in urban areas and inset villages, planning permission

will be granted provided that a proposed development:

“Ensures that the layout, scale, form, massing, height of buildings and structures, and

materials relate to the site context and its surroundings;”…..

There are several other criteria listed and the Policy then goes on to elaborate criteria specific to

inset villages, as follows:

“In addition to the above, proposals for new development within inset village areas will have

particular regard to:

The distinctive settlement pattern of the village and the important relationship between the

built development and the surrounding landscape;

Important views of the village from the surrounding landscape;

Views within the village of local landmarks.”

There are numerous development sites proposed that mean that the first criterion cannot be met, as

the developments proposed do not relate to the scale, form or massing of the site surroundings. For

example, the Horsleys. Here it is proposed that a total of 3.6 km2 should be removed from the green

belt, with a total of 533 houses proposed on sites in East and West Horsley on a total of 29.1 ha, an

average density of 18 dwellings/ha. These proposed developments are not in scale or proportion to

the existing villages, and the housing density is considerably higher than existing housing density.

Effectively, if these developments were to proceed, East and West Horsley would merge and

become a large urban area in the green belt, totally different in character to the existing villages,

with the existing settlement pattern completely lost. Similar considerations apply to other proposed

development sites, such as Flexford and Normandy. In this case, 1.3 km2 would be removed from

the green belt with a total of 1,100 dwellings proposed on 67 ha, an average density of 16.4

dwellings/ha. Again, the scale and form of new development would overwhelm existing hamlets

and villages, merging these to form another urban area in the heart of the green belt, with no

consideration of the views of the settlements from the surrounding countryside, much of which has

been given AONB status.

Thus the words of the policy statement D4 do not match the proposals contained in the draft local

plan.

The “Reasoned Justification” for the policy consists of five paragraphs, 4.5.45 to 4.5.49. Of these,

4.5.45 and 4.5.46, 4.5.48 and 4.5.49 are not justifications for the policy, they are simply statements

or further information. Paragraph 4.5.47 is the only attempted justification for this policy, but in fact

this applies only to insetting. Paragraph 4.5.47 is factually wrong and so the justification for this

policy does not bear close examination. Parts of this paragraph are taken in turn below, to show the

inaccuracies this paragraph contains.

Page 4: Policy D4: Development in urban areas and Inset Villages I

4 | P a g e

“Historically, development has been focused in the urban areas of Guildford, and Ash and Tongham

only. Development in the villages has been very limited due to the Green Belt designation which

previously washed over all but one of the villages.“

This depends on how development is measured. In terms of raw numbers, this may be true, but this

would ignore the fact that urban area of Guildford town contains much of the housing in the

borough. Taking the increase in housing stock between the censuses of 2001 and 2011 in specific

wards reveals a much more complex picture. Effingham, for example, had a higher rate of adding

dwellings than the average for the borough and some of Guildford’s urban wards. Using census data

collected by the ONS, between 2001 and 2011, the number of dwellings in Effingham increased by

6.6%, whereas for the borough as a whole it increased by 5.0%. Over the same period in Merrow, a

ward within the urban area of Guildford, the number of dwellings increased by only 1.2%, whilst in

Normandy ward the increase was 2.6%. This analysis has not been exhaustive, but it is sufficient to

show that the opening statement in paragraph 4.5.47 is incorrect, and that in some cases,

proportionally, considerable development has occurred in green belt villages. It should be

incumbent on GBC to assess the facts before making inaccurate statements such as this. ONS

statistics are open to all.

In the first three months of 2016, planning applications for 12 additional dwellings in Effingham have

been made, in a village of approximately 525 dwellings. Although one of these applications has been

reduced from 6 to 4 dwellings, another 5 have been approved which shows clearly that development

within villages washed over by the green belt can occur, and that organic growth is possible. There

is no necessity for insetting.

“Fifteen villages are now inset from the Green Belt meaning that development is no longer, by

definition, considered inappropriate”.

Guildford’s Green Belt boundaries were established in the Guildford Borough Local Plan 1987, so are

well established. There has not been any justification for the extensive green belt review that

accompanies the insetting of villages, or for the removal of these villages from the green belt.

According to the NPPF, paragraph 83, green belt boundary changes should be made only in

exceptional circumstances, as follows

“Local planning authorities with Green Belts in their area should establish Green Belt boundaries in

their Local Plans which set the framework for Green Belt and settlement policy. Once established,

Green Belt boundaries should only be altered in exceptional circumstances, through the preparation

or review of the Local Plan. At that time, authorities should consider the Green Belt boundaries

having regard to their intended permanence in the long term, so that they should be capable of

enduring beyond the plan period.”

No exceptional circumstances have been demonstrated, with none given as justification in Policy D4

for the policy of insetting. Proposed green belt boundary changes are on a very large scale, as it is

proposed to remove 14.4 km2 from the Green Belt, more than 6% of the existing area designated as

green belt, and to create new green belt boundaries extending to more than 105 km. This is not a

simple revision to the Green Belt, and a demonstration of exceptional circumstances is required for

each and every change. This has not been done, not for a single proposed change.

Page 5: Policy D4: Development in urban areas and Inset Villages I

5 | P a g e

Although Guildford Borough Council are in the process of producing a new Local Plan, this in itself is

not enough to change Green Belt boundaries; the requirement to demonstrate exceptional

circumstances still applies. According to a High Court judgement, Gallagher vs Solihull, issued March

30th, 2014, Case No. CO/17668/2013, there is still a requirement to prove exceptional circumstances

– quotations below are from paragraph 125 of this judgement:

“However, it is not arguable that the mere process of preparing a new local plan could itself be

regarded as an exceptional circumstance justifying an alteration to a Green Belt boundary. National

guidance has always dealt with revisions of the Green Belt in the context of reviews of local plans……

and has always required “exceptional circumstances” to justify a revision. The NPPF makes no change

to this.”

“Exceptional circumstances are required for any revision of the boundary, whether the proposal is to

extend or diminish the Green Belt.”

Whilst each case is fact-sensitive and the question of whether circumstances are exceptional for

these purposes requires an exercise of planning judgment, what is capable of amounting to

exceptional circumstances is a matter of law, and a plan-maker may err in law if he fails to adopt a

lawful approach to exceptional circumstances. Once a Green Belt has been established and approved,

it requires more than general planning concepts to justify an alteration.”

Each and every proposed change to the green belt boundary requires that exceptional circumstances

be demonstrated, otherwise the changes are not lawful. This has not been done by GBC, not for any

proposed change.

From paragraph 130 of this judgement

“In other words, something must have occurred subsequent to the definition of the Green Belt

boundary that justifies a change. The fact that, after the definition of the Green Belt boundary, the

local authority or an inspector may form a different view on where the boundary should lie, however

cogent that view on planning grounds, that cannot of itself constitute an exceptional circumstance

which necessitates and therefore justifies a change.”

“…..That, in his judgment, may now be so: but that falls very far short of the stringent test for

exceptional circumstances that any revision of the Green Belt boundary must satisfy. There is nothing

in this case that suggests that any of the assumptions upon which the Green Belt boundary was set

has proved unfounded, nor has anything occurred since the Green Belt boundary was set that might

justify the redefinition of the boundary.”

No changes that would warrant an alteration to green belt boundaries have been demonstrated by

GBC, their so called “sensitivity analysis” was shallow and subjective, and no changes that may have

occurred since the boundary was established were discussed. The changes proposed do not meet

the standards set in this judgement. Instead, GBC maintain that it is housing need that creates

blanket exceptional circumstances for changes proposed to the green belt. However, this is not the

case. Following a meeting with local councillors and MPs in early 2014, Nick Boles wrote to Anne

Milton MP to clarify the issue of exceptional circumstances and in this letter said:

Page 6: Policy D4: Development in urban areas and Inset Villages I

6 | P a g e

“The written Ministerial Statement of 1 July 2013 set out the Government’s concern that some recent

planning decisions have not accorded the Green Belt the level of protection that was the explicit

policy intent of Ministers. It made clear that the single issue of unmet need – whether for

conventional housing or for travellers sites – is unlikely to outweigh the harm to the Green Belt and

other harm to constitute the “very special circumstances” justifying inappropriate development in the

Green Belt.”

This was reaffirmed more recently in a letter from Brandon Lewis to all English MPs, dated

07/06/2016. In this letter, Brandon Lewis said

“The Government has put in place the strongest protection for the Green Belt. The Framework

makes it clear than inappropriate development may be allowed only where very special

circumstances exist, and that Green belt boundaries should be adjusted only where very special

circumstances exist, through the Local Plan process and with the support of local people. We have

been repeatedly clear that demand for housing alone will not change Green Belt boundaries.”

These quotations show that the insetting process has considerable hurdles to overcome, including as

it does the revision to Green Belt boundaries that accompanies insetting. Thus the most important

questions are – did the flawed Pegasus “Guildford Borough Green belt and Countryside Study” show

unequivocally that villages to be inset are not open villages, making no or a minimal contribution to

the openness of the Green Belt; and that there were exceptional circumstances (other than unmet

housing need) that justify a change to the Green Belt, creating a new boundary around all these

villages – more than 105 km of new boundary? The Pegasus study did not do this, and this was just

one of its many flaws. Some of these flaws are discussed in detail in Annexe 2.

As shown above, development has occurred within villages, and the NPPF made it easier for

additional development to take place within the green belt, via the provisions of paragraph 89,

which specifically makes it easier for limited infill to occur and on a larger scale, for affordable

housing to be built within villages washed over by the Green Belt. These provisions came into effect

in 2012 when the NPPF was adopted and so the rate of increase in housing within villages is very

likely to increase beyond that observed in the period between the 2001 and 2011 censuses.

It is clear from the most recent figures for planning applications to GBC in 2016 in Effingham that the

pace of development within villages is changing, even with continued green belt protection. There is

no requirement for insetting villages within the green belt, and a desire to increase development

within the villages is not an adequate basis for changing green belt boundaries to permit insetting.

The final part of paragraph 4.5.47 is

“In accordance with national policy, the important character of these inset villages should instead be

protected using other development management policies”.

This is not national policy, national planning policy is provided by the NPPF, and paragraph 86 of the

NPPF is the relevant paragraph, which reads

“If it is necessary to prevent development in a village primarily because of the important contribution

which the open character of the village makes to the openness of the Green Belt, the village should

be included in the Green Belt. If, however, the character of the village needs to be protected for other

Page 7: Policy D4: Development in urban areas and Inset Villages I

7 | P a g e

reasons, other means should be used, such as conservation area or normal development

management policies, and the village should be excluded from the Green Belt.”

There is thus no national policy that requires insetting. Instead, a judgement should be made for

each village based on whether the village is open, and if it makes a contribution to the openness of

the Green Belt. In fact, many of the villages proposed for insetting exhibit an open pattern of

development, and make an important contribution to the openness of the green belt, for example,

Effingham. This has not been considered, other than in the flawed Pegasus report, and the

examination of their approach included in Annexe 2 reveals how this analysis was flawed. No

further consideration has been given by GBC to insetting, and so GBC’s policy of wholesale insetting

does not accord with a national policy, real or imagined, as claimed.

In itself, there are many issues with insetting, and these include:

According to the draft Local Plan, insetting is based largely on the findings of the Pegasus

study, “Guildford Borough Green Belt and Countryside Study. This report was very poor and

various parts of the study criticised by many, including some borough councillors. It is

perverse to continue with any recommendations of this study. In particular, a detailed

review of this report for Effingham village showed that it was inaccurate and subjective. This

review is attached as Annexe 2.

The settlement hierarchy was also subject to a detailed review and again this showed that

scoring methods used were inaccurate and subjective.

Proposed boundary changes are arbitrary, without an adequate detailed survey. There are

no checks and balances to this, even though the positioning of the boundary line can change

the price of an acre of land from about £15,000 to over £1 million. There is no process to it,

simply someone using a computer and a mouse – literally a mouse click can turn some land

owners into millionaires. In some cases, the proposed boundary divides semi-detached

houses, so that one is still inside the green belt and the other outside. In other cases, the

boundary divides gardens, so that part of a garden is in, part out of the green belt. Due care

and attention has not been given to producing proposed boundaries.

In many cases, boundaries proposed do not comply with the requirements of the NPPF,

which is given in the NPPF, paragraph 85, final bullet point, which states

“define boundaries clearly, using physical features that are readily recognisable and likely to

be permanent”.

Given this NPPF definition, treelines, hedgerows, woodlands, etc should not be considered

as potential Green Belt boundaries. In fact, in some cases proposed boundaries cross open

fields, with no discernible features whatsoever. These are entirely inappropriate as a green

belt boundary. The boundary proposed for Effingham village is reviewed in Annexe 3.

It should also be noted that GBC implicitly recognise that insetting is likely to lead to more

development within villages on sites that are not identified in the draft Local Plan. But there is no

recognition of this in the housing target. As is noted, insetting would mean that the rate of windfall

development would increase, but GBC have not provided a proper analysis of windfall development

(such as that provided by Mole Valley District Council) and they have failed to recognise a higher rate

Page 8: Policy D4: Development in urban areas and Inset Villages I

8 | P a g e

of windfall development in showing how the housing target will be met – even though, via Policy D4,

they are planning for it. Windfalls should be deducted from the housing target, and the need for

additional sites would thus be substantially reduced. Windfall development should be properly

assessed and identified within the housing target, with an appropriate increase in the rate of

windfalls in any villages that are inset.

Although a flawed justification was provided for the insetting of villages, this does not apply to all

the areas selected for insetting, many of which are not “villages”. Some of these are adjacent to an

Area of Outstanding Natural beauty (AONB) and in at least one case some areas of AONB have been

included in the inset area. These areas include:

o HM Prision, Ripley Road, Send

o Keogh Barracks

o Pirbright Barracks (small areas of AONB included in inset area)

o Pirbright Institute

o Henley Business Park

o Garlick Arch

No attempt has been made to justify the insetting of these sites. In the absence of any attempt at

justification there is no reason to inset these areas.

R. McK rev 1 24/05/06

Submission made by: Roland McKinney

19 Woodlands Road

KT23 4HG

Email: [email protected]

Page 9: Policy D4: Development in urban areas and Inset Villages I

9 | P a g e

Annexe 1

Contribution to Traffic Movements and Air Pollution

Summary

The spatial planning in this draft plan will increase traffic movements and

consequently, air pollution. Air quality should be a constraint used to limit the

housing target. Even with a lower housing target, instead of what is proposed

in the draft Local Plan a higher proportion of housing developments should be

within Guildford town. This could be achieved by recognising that an expansion

in retail and warehousing and distribution within the urban area is not

sustainable, and the land set aside in the town centre for these used instead to

provide housing.

GBC should also have used the local plan to set targets for air quality, with

proposals to reduce noxious emissions, for example, through the use of electric

and LPG fuelled public service vehicles, as is happening in other areas, such as

Birmingham. This has not been done; there is no proposal to tackle existing air

pollution or to alleviate the pollution caused by adding approximately 24,000

cars and vans to those already in use within the borough (based on car

ownership levels in a rural village and an urban area in Guildford). This is the

approximate number of vehicles that the proposed housing additions would

bring, thought his number would vary, dependent on where the additional

houses are located. As an extreme, consider that if all the proposed housing

were within Guildford, then instead of 24,000 about 13,000 cars and vans would

be added to the existing total. In the census of 2011, the total across the

borough was just under 54,000 cars and vans, so if the current plan were to be

implemented there would be a very considerable impact on traffic and air

pollution, and this is lessened considerably by building homes within the urban

area of Guildford.

The comparisons made above show that traffic and air quality has not been

considered in the spatial allocation of housing. Sustainability claims have not

been analysed in drawing up the housing proposals featured in the draft Local

Plan. A simple comparison of only one aspect of sustainability, car use (with its

associated air pollution), shows very clearly that building extra housing in

Effingham (a typical rural area village) is the least sustainable option. As a

method of travel to work, car use in Effingham was shown to be much higher

than in an urban area of Guildford, so to limit increases in traffic and emissions

from traffic, housing should be provided in areas closer to employment

opportunities and with access to good public transport. This does not mean that

no additional housing should be provided in the countryside, but this housing

should be limited in quantity, in keeping with its Green Belt status.

Page 10: Policy D4: Development in urban areas and Inset Villages I

10 | P a g e

Discussion

The Local Plan proposes building a minimum of 693 houses per year, of which

2,400 (17%) will be built within Guildford. The remainder (83%) will be built in

the countryside. This is very much against advice from the Institute for Air

Quality Management. Intuitively, it seems likely that this spatial distribution of

housing would increase car and van journeys with a consequential increase in air

pollution that is related primarily to vehicle emissions. This paper examines

statistical evidence to assess the impact on vehicle journeys within the borough.

Spatial Vision

The spatial “vision” proposed in the Local Plan is the reverse of that

recommended by the Institute for Air Quality Management, who said2

“The pattern of land use determines the need for travel, which is in turn a major

influence on transport related emissions. Decisions made on the allocation of

land use will dictate future emissions, as many people and businesses will make

significant use of road transport for journeys between places that form part of

their daily lives.”

This has not been recognised in the spatial planning within this plan, which

proposes that a large majority of new dwellings should be in rural areas, some

considerable distance from the urban centre and employment centres. It is a

plan that could have been written specifically to increase road traffic.

This report2 also suggested that

“Ideally, air quality should be a prime consideration for long term planning, so

that land is used and allocated in ways that minimise emissions and that reduce

the exposure of people to air pollution.”

This has not been part of spatial planning, but it should have been, even though

there is no statutory requirement to do so. However, there is a statutory

requirement to declare an air quality management area (AQMA) in areas that are

known to have high pollution levels, but this has not been done within the

borough. This is despite requests to the council to set up an AQMA following air

quality tests that show high pollution levels in at least one village which has high

traffic levels.

Sustainability

To examine the claim that the draft Local Plan contributes to sustainability

several areas could be analysed. One of these is the impact on traffic, to assess

the impact of adding additional housing in the countryside rather than in

Guildford centre. This can be gauged by comparing the likely impact on car

2 Land Use Planning & Development Control: Planning for Air Quality, April 2015. Institute for Air Quality

Management

Page 11: Policy D4: Development in urban areas and Inset Villages I

11 | P a g e

ownership of building housing in a rural location as opposed to an urban area of

Guildford. Effingham has been chosen for this, but it could be any of the rural

villages in the borough, and the impact of additional housing in this village is

compared with additional housing in an urban area, Walnut Tree Close/Station

area. This can be done using census details, taken from the Office for National

Statistics (ONS) publications, so that availability of cars and vans can be

compared as well as methods of travelling to work.

What have traffic levels to do with sustainability?

Traffic is a major contributor to air pollution, and air pollution has a significant

effect on health. In April 2014, Public Health England produced a report3 that

reconfirmed the estimate by the Committee on the Medical Effects of Air

Pollutants that approximately 29,000 deaths per year in the UK could be caused

by pollution from man-made particulate matter. This report estimated that

there were 55 deaths per year in the borough attributable to air pollution.

Particulates are very small air borne particles and transport is the single largest

contributor to particulate pollution.

But air pollution is not limited to air borne particles - it includes many other

pollutants, and one group that affects human health, especially that of children,

is the gaseous oxides of nitrogen, nitrogen dioxide (NO2) and nitrogen monoxide

(NO). When taken together these are usually called NOx. Nationally, almost

half of these gases are produced by transport. In a submission to the House of

Commons Report “Action on Air Quality” published in November 2014, Dr Ian

Mudway of King’s College, London said:

“We have also found effects on infant mortality rates, on pre-term birth and on

cognitive performance in children. There is some interesting data emerging on

traffic proximity, diesel emissions and traffic potentially autism spectrum

disorders… The evidence over the last three or four years that children growing

up near traffic in areas with high NO2 and primary particle emissions have

stunted and impaired lung development is incredibly strong.”

As well as health aspects,

road safety is also important –

increased traffic levels means

more congestion, more

delays, more stress, and

inevitably, more accidents.

3 Estimating Local Mortality Burdens Associated with Particulate Air Pollution. See

https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/332854/PHE_CRCE_010.pdf

Page 12: Policy D4: Development in urban areas and Inset Villages I

12 | P a g e

Areas in London exceed the EU limits for various air- borne pollutants, in

particular, NOx. This has been on a scale that will result in the imposition of

substantial fines on the UK government – and the UK government has expressed

a desire for this to be passed to local government in areas that are out of

compliance. Although Guildford Borough has a very limited air pollution

monitoring regime they have one station that regularly exceeds limits – that at

Wisley. Even though this is very poorly sited (meaning that the actual NOx

concentration in this area is considerably higher than that being recorded) it has

consistently been above EU limits. Thus the borough’s taxpayers may have to

share in the costs of the EU fine.

Since traffic has a major impact on air pollution, and as air pollution has a

significant on health, especially that of children, it is an important element in

sustainability – so the impact on traffic levels of housing policies should be

assessed, in detail, before making claims that housing policies are “sustainable”.

Page 13: Policy D4: Development in urban areas and Inset Villages I

13 | P a g e

Detailed Traffic Comparisons

Levels of car ownership vary substantially, as can be seen from the figure, taken

from the data in Table 1. This data is taken from 2011 ONS Census data, as

published in the series “Neighbourhood Statistics”.

The average number of

cars or vans per household

is highest in Effingham,

with an average of 1.9 cars

or vans per household in

Effingham in 2011,

compared to an average of

0.93 cars or vans per

household in Woodbridge

Meadows/Walnut Tree

Close/the Station. This suggests that plans for high levels of development in the

countryside go against all the ideals of achieving sustainable developments as

car and van availability is an indicator of how often those vehicles will be used.

Table 1 Comparison of availability of cars or vans

2011 2001

Effingham (Parish)

Average cars or vans/household 1.89 1.75

Households without cars or vans (%) 4.7 7.3

Woodbridge Meadows, Walnut Tree Close and the Station

Average cars or vans/household 0.93 1.01

Households without cars or vans (%) 34.2 30.0

In Effingham, in 2011, there were very few households that did not have a car

or van available, at 4.7% of total households, compared to 34.2% in

Woodbridge Meadows/ Walnut Tree Close/ the Station. Given the limited range

of shops and services available in rural villages this should not be surprising.

Most residents have to drive to be able to access basic health services, and for

all but the most basic of shopping needs, employment opportunities, etc. This

shows very clearly that it is very likely that there would be increased traffic on

the roads throughout the borough as a consequence of additional housing in the

countryside, and that this increase in traffic would be lower if additional housing

were built in the centre of Guildford.

Experienced city planners such as Lord Rogers have called for increased density

in towns and cities as a way to provide extra housing, provided increased

Page 14: Policy D4: Development in urban areas and Inset Villages I

14 | P a g e

density is accompanied by improved amenities and sensible traffic management

plans. This is because of the benefits of increased density – towns and cities

work better, with better public transport, better access to services, and a

reduction in car journeys. Woodbridge Meadows, Walnut Tree Close and the

Station are ideal locations for increased population density – the railway station

is within walking distance, as is much of the town centre. This is recognised to a

limited extent by the Masterplan prepared for Guildford Borough Council, but

does not seem to have been recognised in the 2016 draft Local Plan.

As the 2011 census had details of the methods used to travel to work, it is

possible to test if traffic implications are in line with these expectations.

Methods of travel to work illustrate differing patterns of car and van use, with

the details given in

Table 2. In Effingham, in 2011, only 14% of employed people who travelled to

work used public transport and almost 75% travelled to work by car or van.

Only 8% of those who travelled to work walk or used a bicycle in Effingham,

whereas in the Woodbridge Meadows/Station area the comparable figure was

almost 33%.

Differences are illustrated in the figure showing the proportion of people who

travelled to work using public transport, walk or by bike. Building more houses

in the countryside will result in many more car journeys – just to get to work.

Many more would also be necessary, for residents to shop, to access financial

and health services, etc. This shows clearly that car journeys would be

minimised if new housing was concentrated in urban areas of Guildford.

Sensible planning should

seek to improve the ability

to use these modes of

transport and the impact

on the number of car or

van journeys is very clear

from the comparisons

made here.

This type of analysis should

be a feature when choosing

to claim an option is

“sustainable”, but it has

not been a feature of the draft Local Plan. Claims made that building housing in

rural areas is a sustainable option do not stand up to simple scrutiny, even

without the consideration that it is proposed to build housing on the limited

resource that is farmland.

Page 15: Policy D4: Development in urban areas and Inset Villages I

15 | P a g e

Table 2 Comparison of Methods Used to Travel to Work in 2011

Travel to Work, 2011 Effingham Woodbridge Meadows, etc

Percentage of those in employment working from home

10.5 5.2

Percentage of those in employment travelling to work by car or van

66.9 34.4

Percentage of those in employment and who travel to work, by car or van

74.7 36.3

Percentage of those in employment travelling to work on foot or by bike

7.1 30.9

Percentage of those in employment and who travel to work, on foot or by bike

8.0 32.6

Percentage of those in employment travelling to work by public transport

13.9 27.4

Percentage of those in employment and who travel to work, by public transport

15.5 28.9

RMck rev 1_22/06/16

By Roland McKinney

19 Woodlands Road

KT23 4HG

Email: [email protected]

Page 16: Policy D4: Development in urban areas and Inset Villages I

16 | P a g e

Annexe 2: Green Belt and Countryside Study Critique

Guildford Borough Green Belt and Countryside Study

by Pegasus Planning Group Ltd

recommended that

Effingham Should be Inset in the Green Belt.

This was not a recommendation that followed objective analysis, as shown in the

review below, which reveals some of the many flaws in this study. Although the

critique uses Effingham village as an example, some of the flaws exposed are of

a general nature, revealing poor quality analysis by Pegasus. These flaws and

other criticisms are applicable to many of the villages considered for insetting by

Pegasus, and so the legitimacy of the entire report is questionable. Because the

report is so seriously flawed it should not form part of the evidence base behind

the draft Local Plan, and any judgements made in the Local Plan that rely on this

report for justification must be considered to be unsound.

Background

In the National Planning Policy Framework (NPPF), paragraph 86 refers to the

criticality of the open character of a village in determining if it should remain

washed over by the Green Belt as follows,

“If it is necessary to prevent development in a village primarily because of the

important contribution which the open character of the village makes to the

openness of the Green Belt, the village should be included in the Green Belt.”

In effect, through their 2013 study Pegasus were assessing if villages within

Guildford Borough make an important contribution to the Green Belt, with

results from that study being used to determine if villages should remain washed

over by the Green Belt, or inset into it. Pegasus concluded that 15 villages

currently within the borough, including Effingham, should be removed from the

Green Belt and inset into it, with new Green Belt boundaries. The proposed

boundary for Effingham is shown in Figure 5 in Annexe 3.

The most important paragraphs of the NPPF that deal with new Green Belt

boundaries are paragraph 83 and 85. In para 83 there is the following:

“…Once established Green Belt boundaries should only be altered in exceptional

circumstances, through the preparation and review of the local Plan….”

Guildford’s Green Belt boundaries were established in the Guildford Borough

Local Plan 1987, so are well established. Although Guildford Borough Council

have embarked on the production of a new Local Plan, this in itself is not enough

Page 17: Policy D4: Development in urban areas and Inset Villages I

17 | P a g e

to change Green Belt boundaries; the requirement to demonstrate exceptional

circumstances still applies. According to a High Court judgement, issued March

30th, 2014, Case No. CO/17668/2013, there is still a requirement to prove

exceptional circumstances – quotations below are from paragraph 125 of this

judgement:

“However, it is not arguable that the mere process of preparing a new local plan

could itself be regarded as an exceptional circumstance justifying an alteration to

a Green Belt boundary. National guidance has always dealt with revisions of the

Green Belt in the context of reviews of local plans…… and has always required

“exceptional circumstances” to justify a revision. The NPPF makes no change to

this.”

“Exceptional circumstances are required for any revision of the boundary,

whether the proposal is to extend or diminish the Green Belt.”

Whilst each case is fact-sensitive and the question of whether circumstances are

exceptional for these purposes requires an exercise of planning judgment, what

is capable of amounting to exceptional circumstances is a matter of law, and a

plan-maker may err in law if he fails to adopt a lawful approach to exceptional

circumstances. Once a Green Belt has been established and approved, it requires

more than general planning concepts to justify an alteration.”

From paragraph 130 of this judgement

“In other words, something must have occurred subsequent to the definition of

the Green Belt boundary that justifies a change. The fact that, after the

definition of the Green Belt boundary, the local authority or an inspector may

form a different view on where the boundary should lie, however cogent that

view on planning grounds, that cannot of itself constitute an exceptional

circumstance which necessitates and therefore justifies a change.”

“…..That, in his judgment, may now be so: but that falls very far short of the

stringent test for exceptional circumstances that any revision of the Green Belt

boundary must satisfy. There is nothing in this case that suggests that any of

the assumptions upon which the Green Belt boundary was set has proved

unfounded, nor has anything occurred since the Green Belt boundary was set

that might justify the redefinition of the boundary.”

Following a meeting with local councillors and MPs in early 2014, Nick Boles

wrote to Anne Milton MP to clarify the issue of exceptional circumstances and in

this letter said:

“The written Ministerial Statement of 1 July 2013 set out the Government’s

concern that some recent planning decisions have not accorded the Green Belt

the level of protection that was the explicit policy intent of Ministers. It made

clear that the single issue of unmet need – whether for conventional housing or

for travellers sites – is unlikely to outweigh the harm to the Green Belt and other

Page 18: Policy D4: Development in urban areas and Inset Villages I

18 | P a g e

harm to constitute the “very special circumstances” justifying inappropriate

development in the Green Belt.”

These show that the insetting process has considerable hurdles to overcome,

including as it does the revision to Green Belt boundaries that accompanies

insetting. Thus the most important questions are – did the Pegasus study show

that Effingham is a not an open village, making no or a minimal contribution to

the openness of the Green Belt; and that there were exceptional circumstances

(other than unmet housing need) that justify a change to the Green Belt,

creating a new boundary around Effingham?

In summary, this review of Volume IV of the Green Belt and Countryside Study

shows that it contained many errors and omissions. This paper deals with errors

discovered that relate to Effingham; specifically, the recommendation to inset

Effingham in the Green Belt as well as the proposed Green Belt boundary around

Effingham. Its conclusions are:

That the recommendation to inset Effingham was based on evidence so

flawed that this recommendation was unsafe,

The proposed Green belt boundary is weak and many sections do not

meet NPPF criteria for new Green Belt boundaries,

No exceptional circumstances justifying a change to the Green Belt

boundary were shown to exist, as is required by planning law,

Effingham should continue to be washed over by the Green Belt.

These considerations probably apply to some or all of the fourteen other villages

that were recommended to be inset into the Green Belt by this study.

Criteria to Assess Villages

At issue is Effingham’s proposed removal from the Green Belt, so that Effingham

would be inset into rather than washed over by the Green Belt. This was one of

the recommendations from Volume IV of the Green Belt and Countryside Study,

2013.

A three stage assessment was developed in the study, called Stage 1, Stage 2,

and Stage 3, and according to Pegasus these were then used

“to test the primary considerations and openness and permanence required

within NPPF paragraphs 85 (last point) and 86 including:

Stage 1: Assessing the degree of openness within each village through analysis

of urban form, density and the extent of developed land;

Page 19: Policy D4: Development in urban areas and Inset Villages I

19 | P a g e

Stage 2: Assessing the locations of new Green Belt boundaries within the

surroundings of each village within Guildford Borough; and

Stage 3: Assessing the suitability of each village for insetting within the Green

Belt and defining potential Green Belt boundaries.”

Each of these Stages is considered below. Relevant maps are given in Annexe 3.

Stage 1: Methodology

In order to assess the openness of villages, Pegasus developed two parameters,

The “extent of perceived village area”

The development footprint within the “perceived village area”

In assessing these Pegasus state that they used site surveys and aerial imagery.

Settlement boundaries were not used to define a village area; instead areas

were selected by Pegasus and were called “extent of perceived village area”.

This was highly subjective, especially since this did not follow actual Local

Authority boundaries that are clearly signposted on main roads and marked on

relevant maps. Consequently, this was a very easy test to manipulate. This was

followed by the derivation of a development footprint which came from computer

analysis of satellite images, but this was not properly assessed and so temporary

wooden agricultural buildings were included in the development footprint.

Finally the ratio of the development footprint to perceived village area was

calculated and expressed as a percentage. Thus the subjectivity of these tests

was masked by adding in some calculations, but the basis of the test was

subjective. Tests of this nature should be developed in an open and transparent

way, so that if other people apply the same test it is highly likely that they would

produce the same numbers. This was not the case for these tests. As examples

of the lack of transparency, no independently defined boundaries were used in

determining either “developed” land or “open” land; this was just an

interpretation by Pegasus – which did not even follow clearly defined Local

Authority boundaries.

Following the calculation of the ratio of development footprint to perceived

village area, the classification of villages using this statistic would be expected to

follow some sort of pattern. In this case, the range of values was about 10% to

25%. Thus a typical classification would be something that divided the range

into three equal portions – low would be below 15, medium 15 to 20, and high

anything above 20. This approach was not followed, and so there was no logic

to Pegasus’s classification – in other words, although numerical tests were

devised by Pegasus they ignored the numerical results. This being the case,

what was the point in developing numerical tests?

Page 20: Policy D4: Development in urban areas and Inset Villages I

20 | P a g e

Stage 1: Application to Effingham

In the specific case of Effingham “the perceived village area” extended into the

Mole Valley District Council (MVDC) region, with all of Woodlands Road and some

developments along Guildford Road included – most of which are in Mole Valley.

However, the open fields (Rolls Farm) opposite the entrance to Woodlands Road

were not included in “the perceived

village area”, whereas all the housing

was included. It seems it was

acceptable to include developed areas

from Mole Valley, but not open land,

suggesting that the purpose behind

the selection of the “extent of

perceived village area” was to get a

predetermined answer. In this

satellite image of this area, Rolls

Farm, the excluded open area is

marked – but all the housing shown

was included, including Dobbes

nursery. Along the A246 (travelling

towards Effingham from Bookham) Effingham village signs are located after the

entrance to Woodlands Road but just before the junction with Manor House

Lane; and if travelling in the opposite direction there is a Little Bookham sign

after Manor House Lane and before the entrance to Woodlands Road. So the

“perceived village area” used by Pegasus could only be done by ignoring these

very visible boundary limit signs. Thus the test developed by Pegasus was

manipulated to provide a pre-determined answer; it was not an objective

examination of facts, as it should have been. Housing within Mole Valley should

not have been included in the “perceived village area”.

The next stage was to assess the proportion of development within this

“perceived village area”. In the derivation of the developed or built area, the

built area was measured by computer analysis of satellite images. An

examination of just a small area showed that this computer based analysis

provided flawed data.

Page 21: Policy D4: Development in urban areas and Inset Villages I

21 | P a g e

Temporary agricultural buildings, including wooden stables, at the top of

Woodlands Road were included in the “built” total - two of these are identified in

the satellite image, above. One of these is wooden field stabling for horses, and

so should not be included in a developed or built area total. In addition, both

are in Mole Valley, not Guildford Borough. Errors represented by these probably

occur across the rest of the area, with other temporary agricultural buildings and

possibly large garden sheds (in Mole Valley as well as Effingham) included in the

“built area”. These errors were probably repeated in their analysis of other

villages, thus the “built” or developed area as measured by Pegasus was an

unreliable statistic. (See Stage 1 map, in Annexe 3).

These two numbers – “perceived village area” and “built area” were then used

by Pegasus in their two part assessment of the openness of villages, Stage 1.

Taking the first criterion developed by Pegasus, the quotation below was taken

from the Pegasus Report, Volume IV, section 14.17:

“Does the majority of the village exhibit an open character?

Effingham generally exhibits a high proportion of built development to open land

with approximately 13.87% of the village area (160.33 ha) covered with existing

development (22.24 ha)”

The derivation of this statistic, (22.4 x 100/160.33 = 13.87%) was questionable,

as detailed above. In this test, lower numbers indicate lower levels of

development within a village, and so in this case the interpretation of “open” was

land that has not been developed.

The use of two decimal places was foolish. To a numerate person, reporting

numbers to two decimal places indicates a very high level of confidence in the

numbers used, but as discussed, the methodology used to create them was

subjective. Other people repeating this test would be unlikely to create the

same value for “perceived village area”, simply by using the visible village

Page 22: Policy D4: Development in urban areas and Inset Villages I

22 | P a g e

boundary signs. Equally, anyone else using computer analysis of the built area

may choose to check results – and then remove anomalies such as the inclusion

of temporary agricultural buildings, etc, thus giving an entirely different number

for development footprint. So a high level of confidence in this number was

misplaced and reporting these numbers to two decimal places was highly

inappropriate and misleading.

Furthermore, the interpretation of the calculated ratio by Pegasus was entirely

without logic. This ratio, expressed as a percentage, was given for 24 villages

across the borough and the following is a summary of those ratios (as

percentages) within the classifications for individual villages given by Pegasus:

“Low” category:

14 villages in this category, with values ranging from 9.7% to 21.6%

“Medium” category:

1 village in this category – 17.1%

“High” category:

9 villages in this category, with values ranging from 13.9% to 24.8%

Clearly, there was no consistency in this classification by Pegasus, with a very

considerable overlap between villages categorised as “low” and those

categorised as “high” with only one classed as “medium”.

What was key for Effingham was that it was incorrectly categorised as “high”,

with a score of 13.9%, when “medium” was given as 17.1%. In other words, if

there was any consistency in this approach by Pegasus, Effingham would have

been categorised as having a low development footprint to open land ratio. And

this ratio was obtained even after selective inclusion of developed areas from

Mole Valley and wooden agricultural buildings in the development footprint. But

the actual calculated ratio was ignored, suggesting that Pegasus was

predisposed to the insetting of villages before they even started their analysis.

There was a second part to Stage 1, and the next section deals with this -

consider another quotation from the Pegasus report (section 14.17):

“Effingham generally exhibits a high density of development…”

No numerical definition of high, medium, or low densities was given though this

would have been appropriate. Three developed areas within the perceived

village area were given as examples of high density – out of fourteen Effingham

village areas considered. Four of these were described as either open farmland

or low density and the remainder as medium density; so even on the basis of

Page 23: Policy D4: Development in urban areas and Inset Villages I

23 | P a g e

their non-numerical analysis this was an astonishing conclusion – three out of

fourteen areas defined the overall density of the village!

Areas categorised as high density included an area designated 1A on the Stage 1

Effingham map (see Annexe 3). This is an area of 6 houses, set in 1.73 acres

(0.7 ha) so the calculated density is 3.5 houses/acre, or 8.6 dwellings/ha.

Under no circumstances could this be considered as high density.

Area 1C was described as high density, but the actual density is 12.2 houses/ha.

Area 1K, described as high density, actual density is 17.9 houses/ha.

These are the only areas described as high density in the Effingham analysis –

for details see maps in Annexe 3 to this critique. To recap, 3 areas were

reported as high density, but actual numerical analysis showed this was wrong.

Setting this flawed analysis aside, 7 of the 14 were described as medium density

and 4 as low density or open farmland. Yet the conclusion was that the three

areas (erroneously) described as high density typified all of Effingham, so that

overall Effingham was considered to “exhibit a high density of development”.

There was no logic in this conclusion – and as shown above, the 3 areas

described as high density should not have been described as high density.

Again, this is indicative of Pegasus having a predetermined result in mind when

performing the tests.

By way of comparison, in the Guildford Borough Council Monitoring Report

2013/14, published in June 2014, the average density of new developments

granted planning permission in 2013/14 was 111 dwellings/ha in Guildford town

centre and in villages was 71 dwellings/ha. These are much higher densities

than anything in Effingham. Consider too that the average density of one of the

proposals from Berkeley Homes to replace the Howard of Effingham school was

25 dwellings/ha; with the highest proposed density (on the existing school

playing fields, running into MVDC, along Lower Road) at 31.2 dwellings/ha. This

was described by Berkeley Homes as “low density”. So to suggest any of the

areas within Effingham considered in the Pegasus report were high density

development was simply wrong. As Pegasus did not provide any numerical

definitions for “high”, “medium” or “low” or provide any analysis to support their

observations it is more difficult to challenge their conclusion. However, the

numbers from a proper numerical analysis of actual housing density speak for

themselves – Effingham generally has a low density of housing.

Stage 2: Assessing the Locations of Potential Green Belt Boundaries

According to their report, Pegasus carried out detailed analyses to identify

potential Green Belt boundaries, using site surveys, aerial imagery and detailed

Ordnance Survey mapping. However, this was used to map woodlands,

Page 24: Policy D4: Development in urban areas and Inset Villages I

24 | P a g e

hedgerows, treebelts highway and railway infrastructure as potential Green Belt

boundaries. They quoted the NPPF paragraph 83 which says local authorities

“….should consider the Green Belt boundaries having regard to their intended

permanence in the long term, so they should be capable of enduring beyond the

plan period.”

However, they ignored NPPF paragraph 85, final bullet point, which states

“define boundaries clearly, using physical features that are readily

recognisable and likely to be permanent”

Given this NPPF definition, treelines, hedgerows, woodlands, etc should not be

considered as potential Green Belt boundaries. Pegasus appeared to have

simply ignored this paragraph, even though it was quoted. Treelines, woods and

hedgerows are not physical features but are biological features. As biological

features they are subject to disease, old age, windthrow and of course removal

by man. Diseases are widespread – many horse chestnut trees in Effingham are

clearly diseased, and started to shed their leaves in August. Ash die back is

becoming more widespread and is present in Effingham. Elm trees were all but

wiped out by Dutch elm disease: so whilst treelines etc provide landscape

features they cannot be considered to be permanent. The Woodland Trust, for

example, reports that more than 300 ancient woodlands are currently under

threat from development. Development is a greater enemy of woodlands than

all other threats combined. Consequently, woodlands, treelines, etc do not meet

NPPF Green Belt boundary criteria.

Pegasus mapped detailed possible Green Belt boundaries around all villages,

without considering if exceptional circumstances had been demonstrated that

would justify a change in the Green Belt boundary. Detailed potential

boundaries are given for each village in Volume IV of the Pegasus report, but

these do not necessarily correspond exactly to the proposed new Green Belt

boundaries around the villages that GBC propose to inset into the Green Belt,

although in many cases a large proportion of the boundary perimeter is based on

recommendations made by Pegasus.

Application of Stage 2 to Effingham

In their mapping process, Pegasus actually suggested Green Belt boundaries for

the village using features that are in Mole Valley. They listed eighteen features

that they suggested could form new Green Belt boundaries, but four of these

were woodlands in Mole Valley (see Stage 2 map in annexe 3). It would appear

that during their detailed mapping process Pegasus were unable to recognise

local authority boundaries on their maps. Only three of the eighteen actually

followed physical features – roads. Of the remainder, two were fencelines, and

the rest woodland features of some description. However, despite their

Page 25: Policy D4: Development in urban areas and Inset Villages I

25 | P a g e

unsuitability, many of the boundaries proposed by Pegasus feature in the

proposed new settlement boundary perimeter for Effingham, including two

fencelines - which are wholly unsuitable. Other new sections to this were

proposed as new settlement and new Green Belt boundaries in the draft Local

Plan. New combined Green Belt and settlement boundaries for Effingham are

considered in more detail in Annexe 3.

Stage 3: Assessing the Suitability of Each Village for Insetting within

the Green Belt and defining new Green belt Boundaries

This is the stage that provides the insetting criteria which were given as

“Does the majority of the village exhibit open character?

Do open areas within the village generally appear continuous with

surrounding land beyond the village – from within/or outside the village?

Does the majority of the village exhibit incomplete, indistinguishable

boundaries that would not permit the provision of New Green Belt

boundaries in accordance with the requirements of NPPF paragraph 85

(last point)?”

Under stage 3, insetting criteria were developed which were then evaluated (in

tabular form) for each village. Each village was given a positive or negative

against each criterion – all very subjective. If the results of all three stages

were positive, positive, positive then it was considered that the village would not

be considered appropriate for insetting within the Green Belt.

Although NPPF paragraph 85 was referred to, it was ignored; even though this

forms the core of the requirements for new Green Belt boundaries, as was the

need to demonstrate “exceptional circumstances” when changes are proposed to

established new Green Belt boundaries. This latter point was not considered by

Pegasus in their insetting criteria, despite being crucial to the result. If no

exceptional circumstances can be demonstrated, then new Green Belt

boundaries cannot be created, so no insetting would be possible.

Pegasus then claim to have made “professional judgements” based on these

tests – in other words, entirely subjective judgements that are not open and

transparent. They gave positive or negative responses for each one of the three

insetting criteria for each village.

The first of these insetting criteria, openness, related closely to Stage 1 analysis,

but the openness of the village, was scored as negative by Pegasus. However,

Page 26: Policy D4: Development in urban areas and Inset Villages I

26 | P a g e

Pegasus were wholly inconsistent in the way they applied and interpreted these

tests. Without their errors this test would have been a positive for Effingham.

There was considerable ambiguity in interpreting the second insetting criterion

“Do open areas within the village appear continuous with surrounding open land

beyond the village – from within and/or outside of the village?”

There is a relationship between this and the first test, the openness of the

village. In summarising the negative for openness attributed to Effingham,

Pegasus said

“On balance, the majority of the village is considered to exhibit an enclosed

character.”

Where villages were shown to have a high development footprint (covered in

Stage 1 analysis) it would be likely that there would be few open areas. In fact,

Pegasus gave Effingham a positive value for this second criterion, which to some

extent contradicts their award of a negative score for their openness test, and

especially contradicts the line quoted above; in which Pegasus suggested

Effingham was an enclosed village. It then seems odd that an adequate number

of open areas were found to be continuous with surrounding open land, to the

extent that the award of a positive was justified when they said “the majority of

the village is considered to exhibit an enclosed character.”

All roads through Effingham show how open the village is, and the contribution

made by the village to the openness of the Green Belt. Pegasus did consider

that some views were obstructed by woodland – and that these views prevent a

connection with the wider Green Belt beyond. Taken to its logical conclusion,

this would have all those campaigning to keep their village in the Green Belt

removing trees in and around their village. The absurdity of this is beyond

words. In any case, the degree of openness is affected seasonally – when trees

lose their leaf cover (about half of the year) views are much more open, so

campaigners should not feel any need to reach for their chainsaws.

Some views of the Green Belt from roads and footpaths in Effingham:

Page 27: Policy D4: Development in urban areas and Inset Villages I

27 | P a g e

From the A246

From Lower Road

From Effingham Common Road

Page 28: Policy D4: Development in urban areas and Inset Villages I

28 | P a g e

From Effingham centre

All of these views reflect the openness of the village and connections with the

wider Green Belt beyond the village. However, photographs do not adequately

convey this openness – many other views could be shown, to illustrate this

openness. Pegasus’s approach to the assessment of the openness of Effingham

was inadequate, in that they claimed connections to the wider Green Belt were

intermittent, whereas the reality is that Effingham is open, and connections to

the wider Green Belt are easily made from many locations within the village.

Even though Pegasus considered open views were intermittent, they did award a

positive score for this test, which was the correct outcome.

The third insetting criterion (part of the Stage 3 analysis) was given as

“Does the majority of the village exhibit incomplete, indistinguishable boundaries

that would not permit the provision of new Green Belt boundaries in accordance

with the requirements of NPPF paragraph 85 (last point).”

The National Planning Policy Framework (NPPF) bullet point in paragraph 85

referred to reads:

“define boundaries clearly, using physical features that are readily

recognisable and likely to be permanent”

This was discussed under Stage 2 tests and more details are given in Annexe 3.

The Green Belt boundary suggested by Pegasus for Effingham is shown on the

Stage 3 map in Annexe 3. As this is very different from the Green Belt boundary

proposed in the 2016 GBC draft Local Plan it is not discussed in depth, though

both proposals share some sections of the boundary. In summary, only

relatively small sections of the Pegasus proposed Effingham Green Belt boundary

were based on physical boundaries likely to be permanent. Other sections used

features such as woodlands, which are not physical, nor are they likely to be

permanent. Two sections were based on fencelines, which were wholly

Page 29: Policy D4: Development in urban areas and Inset Villages I

29 | P a g e

inappropriate as a new Green Belt boundary. Another two sections have no

feature of any description, the proposed boundary crossed open green fields.

Pegasus arrived at an incorrect conclusion by suggesting that these boundaries

would meet NPPF requirements. Consequently, the negative score given to

Effingham should be reversed to a positive score. In addition, no extraordinary

circumstances were demonstrated, so new Green Belt boundaries cannot be

created.

If the scoring of insetting criteria had been applied consistently and in

compliance with the NPPF, Effingham would have been scored positive in all 3

insetting tests and so would not be considered for insetting.

Tests of this nature should be developed using methodology that is transparent

and repeatable, so that similar results would be generated by others repeating

the tests. Methodology developed by Pegasus did not meet these requirements

as methodology used in all evaluation stages was subjective and so open to

exploitation. This was not acceptable for tests of this nature. Possibly the most

important aspect was not even considered – the need for exceptional

circumstances to be demonstrated before Green Belt boundaries can be

changed. Consequently, results of the evaluation of insetting of villages within

GBC should be disregarded.

By: Roland McKinney

19 Woodlands Road

KT23 4HG

Email: [email protected]

Page 30: Policy D4: Development in urban areas and Inset Villages I

30 | P a g e

Annexe 3

Proposed New Effingham Green Belt Boundary

A new Green Belt boundary for Effingham village has been proposed in the GBC

2016 draft Local Plan, shown in Figure 5. There have been other maps proposed

for a green belt boundary around Effingham village, one in the previous (2014)

draft Local Plan and another by Pegasus. This discussion refers to the 2016

draft Local Plan map and is relevant to the decision on whether or not Effingham

village should be inset into the Green Belt.

The proposed new green belt boundary – which will become the de facto village

settlement boundary is shown as a green dashed line in Figure 5. As the

proposal is to inset Effingham village into the Green Belt this proposed new

boundary would also be an entirely new Green Belt (GB) boundary as Effingham

village is currently washed over by the Green Belt and so has no boundary with

it. According to the NPPF, the GB boundary can be changed only in exceptional

circumstances, but no exceptional circumstances have been presented to

warrant this change. In addition, the NPPF specifies (para 85) that

“…when planning boundaries, local planning authorities should:…….

define boundaries clearly, using physical features that are readily

recognisable and likely to be permanent”.

There are many sections of the proposed boundary that do not meet this

criterion, with some sections having no discernible visible feature of any

description. Other sections have very temporary markers, such as a simple fence

line. One of these is illustrated in the satellite photograph below, Figure 1, part

of the section between the A246 and Orestan Lane, adjacent to The Grove, with

the section of the proposed green belt boundary marked by dashed red line.

This boundary is provided by a simple wooden fence line, with no permanent

features of any description marking the proposed Green Belt boundary. It does

not meet NPPF guidelines for physical, permanent features.

The continuation of this of this section is illustrated in Figure 2, and again there

are no permanent physical features.

The final part of this section is shown in Figure 3, and again there are no

permanent physical features – this section is essentially a broken tree and fence

line. This entire section, from the A246 to Orestan lane does not meet the

requirements of the NPPF, and so is unsuitable for a green Belt boundary.

This is not the only section that is not suitable to be a boundary, there are many

others, such as the section around Leewood Park, the section behind Strathcona

Avenue from woodlands Road, and parts of the section from Lower Road to the

A246.

Page 31: Policy D4: Development in urban areas and Inset Villages I

31 | P a g e

Figure 1 Proposed G B Boundary – A246 to Orestan Lane (The Grove)

Figure 2 Proposed Green Belt Boundary – The Allotments

Figure 3 Final Section of Proposed Boundary to Orestan lane

Page 32: Policy D4: Development in urban areas and Inset Villages I

32 | P a g e

Part of this latter section is marked by a footpath, which is a recognisable

boundary, but one part crosses the Howard of Effingham school playing fields.

There is no boundary marker in this area, though there are a few trees and

some hedging – but some is just an open green field. This section is illustrated

in Figure 4 – this is one of the weakest sections of the proposed Green Belt

boundary.

Figure 4 Proposed G.B. Boundary – HOE School Playing Fields

As many sections of the proposed GB boundary do not meet the requirements of

the NPPF for new Green Belt boundaries, Effingham cannot be inset from the

Green Belt. Where there are features, many of these are not physical features,

just biological – tree lines and hedging, or wooden fences. Treelines in particular

are subject to wind damage, disease, and removal by land owners. These

simply cannot be considered to be permanent physical features. This is entirely

separate from the need to show exceptional circumstances in order to change

the boundary, even during a process to produce a Local Plan. A review of the

Green belt boundary during the Local Plan process in itself is not sufficient to

change the boundary; exceptional circumstances still have to be demonstrated.

In addition, it is necessary to show what has changed since the original Green

belt boundary was established.

These photos serve to illustrate that sections of the proposed new Green Belt

boundary around Effingham do not meet NPPF criteria for new Green Belt

boundaries. In the absence of exceptional circumstances and suitable

boundaries Effingham cannot be inset into the Green Belt. These conclusions

are likely to apply to many other villages that are proposed to be inset by the

2016 draft Local Plan.

Page 33: Policy D4: Development in urban areas and Inset Villages I

33 | P a g e

MAPS

Figure 5: GBC Proposed New Effingham Settlement and Green Belt Boundary (2016)

Page 34: Policy D4: Development in urban areas and Inset Villages I

34 | P a g e

Stage 1: Assessing the Openness in Effingham, Analysis of Village Form, Density and Extent of Developed Land (from Pegasus Study)

Page 35: Policy D4: Development in urban areas and Inset Villages I

35 | P a g e

Stage 2: Assessing the Defensible Boundaries for Green Belt Surrounding Effingham (from Pegasus Study)

Page 36: Policy D4: Development in urban areas and Inset Villages I

36 | P a g e

Stage 3: Assessing the Suitability of Insetting Effingham within the Green Belt and Defining new Green Belt Boundaries (Pegasus study)

Roland McKinney 07/07/2014: rev 1

Page 37: Policy D4: Development in urban areas and Inset Villages I

37 | P a g e

By: Roland McKinney

19 Woodlands Road

KT23 4HG

Email: [email protected]