Proposed Main Modifications

1. Examination of Evidence Document ED159 - Note to the Inspector: Implications of new housing projections for NHDC Local Plan

THE REPRESENTOR'S ARGUMENT
Introduction:
In the light of the new housing projection statistics issued by the Office for National Statistics in late 2018 and which seem to indicate a lower projection for housing requirements across the UK, the Inspector asked North Hertfordshire District Council to comment on whether there are any implications arising from these new numbers generally and, specifically, in relation to the Local Plan's proposal to meet unmet housing need from Luton and on land currently designated as Green Belt around Cockernhoe. NHDC's response is contained in Evidence Document ED159.
Looking at the NHDC paper ED159 and the statistics contained within and cross-referencing that to other documentation from, specifically, Luton Borough Council and Central Bedfordshire District Council it is very apparent that, from almost all methods of analysis, there is in fact NO UNMET HOUSING NEED FROM LUTON.
ED159, and with it the requirement from North Hertfordshire District Council to build 2,100 homes in the area around Cockernhoe, should therefore be rejected as unsound.

The simple analysis
According to the adopted Luton Local Plan the number of homes required in the town itself has been identified as 17,800 during the period 2011-2031. Luton Borough Council has committed to building at least 8,500 homes within the Borough over the 2011-2031 period and this number is committed to in the Luton Local Plan. The difference between the required housing in Luton of 17,800 and the planned building of 8,500, results in a shortfall - the so-called unmet housing need from Luton - of (17,800-8,500) = 9,300 homes.
Of this total of 9,300 homes, Central Bedfordshire DC has committed to build 7,350 homes in its area adjacent to Luton while NHDC has offered to build the balance of 1,950 homes in the Green Belt area around Cockenhoe.
However, the new ONS 2016 housing statistics, which predicts fewer houses being required across the UK than previously estimated, shows a reduction of 519 homes per year to be required in Luton. Measured over the 20-year period of the Local Plan that results in 10,380 fewer homes being required in Luton.
Therefore, an initial simple analysis shows that the figure of 10,380 fewer homes wipes out Luton's unmet housing need of 9,300 homes, by a margin of 1,080 homes. Looking at the statistics another way, if the new ONS statistics suggest that Luton's housing needs are not now 17,800 but 17,800 -10,380 = 7,420, then the planned building of Luton's 8,500 homes comfortably exceeds the number of homes required and, thus, no unmet housing need exists.

Looking at this is more depth

NHDC's ED159 document makes considerable mention of the uncertainties surrounding these new ONS statistical projections. Their response includes papers quoting a senior UK government figure citing reasons as to why there may be anomalies in these statistics. However, the reasons given for why the new statistics (which were compiled based on trends evident between 2001-2011) might be subject to some doubt, mention several key pointers including low levels of housebuilding in that decade, a decline in housing affordability, and, increasing numbers of adults living at home for longer. We would argue that those pointers are still in evidence today and that, therefore with these underlying pointers still valid today, it is quite conceivable that the new ONS statistics might well be accurate.
However, we have to recognise that Government guidance to local authorities states that the earlier ONS statistics should still be used for the examination of Local Plans that have already been submitted.
Under the Duty to Co-operate one would have thought that NHDC would have taken the trouble to look closely at Luton's record in building new homes. So we have done it for them!
If we look at what Luton Borough Council has achieved in terms of building new homes over the past 27 years (that is the previous Local Plan 1991-2011) and the first seven years of the current Local (2011-2018) Luton has achieved an over-delivery of housing averaging out at 56 homes per year over the 27-year period. In fact, in the current Luton Local Plan from 2011-2018, the over-delivery of homes has been 103 homes per year over the past seven years.
Nowhere in its Local Plan submission does NHDC take into consideration this over-building achievement over an extended period by Luton Borough Council and made an effort to extrapolate it to future house building needs in the town.
At the current rate of build from 2011-2018 Luton had built 3,698 homes compared to a plan target of 2,975 in that period (425 homes a year). If this rate of over-build continues for the remaining 13 years of the Luton Local Plan it would result in a total up to 2031 of the already built 3,698 plus the targeted build of 13 years x 425 homes per year = 5,525, plus the over delivery at the current rate of 13 x 103 homes = 1,339 homes. A grand total of 3,698 + 5525+ 1,330 = 10,562 homes, exceeding the number in the Luton Local Plan by 10,562-8,500 = 2,062 homes.
More realistically we have also looked at a scenario where the over-build of homes is limited to the average achieved over the past 27 years - 56 extra homes a year. Under this scenario, the over-build for the remaining 13 years of the current Local Plan would be 3,698 + 5,525 plus 728 = 9,951 new homes, exceeding the number in the Luton Local Plan by 1,451 homes.
Now, taking a look back at the housing shortfall as suggested by the new ONS statistics of 10,380 homes, compared to the 17,800 homes identified by Luton as new homes required during the 2011-2031 period, that leaves only 7,420 homes left to be built.
Luton has already built 3,698 of these leaving 3,722 to be built during the next 13 years - an average of 286 a year. At the lower average rate of over-build of an extra 56 homes per year, that would result in a yearly total of the planned 425 plus the extra 56 = 481 homes per year delivered which, over the remaining 13 years of the Luton Local Plan, equals 6,253 homes which added to the 3,698 already built gives a total build of 9,951 new homes.
Of course, at the current rate of overbuild of 103 extra homes per year that total of 9,951 new homes would grow to 10,562 new homes delivered.
Both of these projected build rates would comfortably exceed the Luton housing figures of 7,420 suggested by the new ONS statistics by anything between 3,412 (the higher rate of build) or 2,531 (the lower rate of build). Thus, even if the new ONS statistics were too pessimistic, and more homes were likely to be needed, the chances are very good that Luton would be able to deliver those extra homes using its own resources and with no unmet housing need arising from Luton.
But, in another scenario, if local authorities are forced to rely on the previous ONS statistics and that, therefore, there is still an unmet need of 9,300 homes arising from within Luton, the fact remains that Luton's record of over-achievement in building homes over the past 27 years would indicate that it could deliver between 2,062 extra homes (higher build rate) or 1,451 extra homes (lower build rate). This would mean that the unmet need of 9,300 would reduce to between 7,238 (higher rate) and 7,850 (lower rate) homes needed.
Central Bedfordshire, in its Local Plan submission, has committed to deliver 7,350 homes to meet Luton's unmet housing needs. At the higher rate of Luton build this would mean that Central Bedfordshire's allocation would meet the Luton unmet housing need; at the lower rate of Luton build only some extra 500 homes would still be needed.
The big question, therefore, is which local authority is to meet that possible Luton unmet housing need.
We argue that Central Bedfordshire is by far the biggest component outside the Luton Borough to meet this need. Its population in the Luton Housing Market Area (HMA) contributes 35.6% of the population of the Luton HMA dwarfing the area around Cockernhoe which contributes a statistically insignificant 0.15%.
Central Bedfordshire is better placed geographically and with better access to superior transport links, key services and infrastructure and closer to the Luton jobs market, to contribute their 7,350 homes first. At the higher rate of Luton over-build this would remove any residual unmet need arising from within Luton. At the lower Luton over- build rate an extra 500 homes would be needed to meet this shortfall.
In this instance it is strongly arguable that Central Bedfordshire has more land identified for possible housing sites (some 266 sites in the Luton catchment area, quite a few of which are not in the Green Belt).
Also, Luton Borough Council has stated that it strongly desires that any housing provided from outside its area to meet any unmet need has to be closer to facilities, noting in particular that it is urging Central Bedfordshire to bring forward the potential of land to the west of Luton for development much more quickly; a subject of some dispute between the two authorities.

A closer look at the statistics contained in the NHDC ED159 response paper in relation to Luton's unmet housing needs
In its ED159 response NHDC has resorted to some strange statistical gymnastics which we have analysed.
First of all NHDC refers to the new ONS statistics as suggesting that Luton's housing needs are some 40% less than the previous ONS projection with some 9,500 fewer homes being required.
But in the statistics tabulation close by in ED159 they appear to contradict their own statement. They correctly list Luton's current plan requirement for new homes as the 17,800 discussed above, but instead of showing the large reduction down to 9,500 for the new indicative figures based on the new ONS statistics, they show the new Luton requirement as being 17,000 homes. For some reason the 40% reduction in homes suggested by the new ONS statistics has become only a 4.9% reduction to 17,000.
NHDC's justification is that they have applied 'vacancy rates and market signals uplifts applied in line with the approach taken in respective Housing Market Assessments.' But there is no particular explanation as to how applying those parameters results in such a potentially large drop in housing needs being reduced by only a fraction.
Looking at Market Signals Uplifts as used by neighbouring authorities we can see that the uplift in housing numbers averages from 5% - 20% (Examples include South West Herts comprising Dacorum, Hertsmere, St Albans, Watford and Three Rivers - 5.1%; Welwyn and Hatfield - 20%; West London Boroughs Alliance - 18%; Tower Hamlets - 20%).
The proposed uplift suggested by NHDC comes out as 79% (an increase from the figure of 9,500 to the suggested figure of 17,000). It is very difficult to see how such a huge market signals uplift can be justified.
But there is more.
In ED159 NHDC provides another tabulation which uses a different housing forecast method based upon Lichfields - a planning consultancy. Using this approach, the housing figures for Luton are reduced from the 17,800 in the Luton Local Plan to 13,000, a reduction of 4,800.
If those Lichfields figures are adopted it means that Luton's plans to build 8,500 homes in the 2011-2031 period will result in an unmet need of 4,500 homes (compared to 9,300 in the Luton Local Plan). It is entirely likely that Luton will continue to over-delivery on its housing build over this period, so this unmet housing need could come down to more like 4,500 - 2,062 (higher build rate) = 2,438 unmet need, or, (at the lower build rate) it could be 4,500 -1,451 = 3,049 as the unmet need.
So, in this scenario, we come again to which local authority is better placed to make up this shortfall. We continue to argue that Central Bedfordshire is far better placed to meet this requirement with its commitment to build 7,350 homes to meet unmet needs from within Luton and so removing the need for NHDC to build 1,950 homes on the Green Belt around Cockernhoe.

But NHDC also comments in ED159 on housing requirements in the part of Central Bedfordshire that falls within the Luton HMA

NHDC then goes on to suggest in ED159 that the area of Central Bedfordshire that falls within the Luton HMA will experience an increase in homes required under the new ONS statistics from 13,400 to 16,000. Again, there is no explanation as to why there is an increase, though it might be due to around a 20% market signals uplift.
But then NHDC applies the Lichfield consultancy methodology and comes up with an even higher increase in homes required for Central Bedfordshire within the Luton HMA from the 16,000 to 20,100 - again no explanation or justification for this increase.
This is despite the fact that, according to the new ONS statistics, Central Bedfordshire will require an average of 238 fewer homes per year - which over a 20-year period equates to 4,760 fewer homes. Indeed, Central Bedfordshire itself states in its response to the UK government consultation on the new ONS statistics that these new ONS statistics would lead to a 'housing reduction for many authorities, including Central Bedfordshire'. So where NHDC gets these increased housing figures from is mysterious.
This is vital because it is clear that NHDC is in the process of trying to manipulate statistics to justify a change in its approach. It is quite clear that they are moving the goalposts.
Up until this ED159 paper is has been abundantly clear that NHDC's offer/desire to build 1,950 homes around Cockernhoe is to meet Luton's unmet housing need - that is from within the Luton town borough area. They now say it is necessary to review the new ONS statistics across the wider Luton HMA.
So we now look across the wider Luton HMA. Taking NHDC's own figures in ED159 we can combine this increase in housing for Central Bedfordshire, with the reduction in the Luton Borough requirement, together with a statistically insignificant increase in Aylesbury Vale from 400 - 600, and leaving 200 homes (as the even more statistically insignificant figure) for North Hertfordshire, to leave the wider Luton HMA as requiring an increase from 31,800 homes to a new higher figure of 33,900 homes - approximately 7% higher NHDC states.
Central Bedfordshire, in its MoU with NHDC in September/October 2018, has said in relation to its own requirement to build 39,350 in Central Bedfordshire, (32,000 for its own needs and 7,350 to meet unmet needs from Luton), that it is currently planning to meet its housing needs in full within the plan period and has confirmed:' that at the current time assistance is not required from any other local authority, including NHDC, to meet its OAN.'
The increase from 32,000 identified by Central Bedfordshire as its housing requirement to the 33,900 suggested by the NHDC figures in ED159 does not seem to be hugely significant (NB no response has yet been found from Central Bedfordshire to these NHDC statistics).
So to conclude this part of the representation, if NHDC's own analysis of Luton's housing needs using the Lichfield methodology is followed, then, at the very most, the Luton unmet need will be 4,500 homes (though with the strong likelihood that Luton will build more homes). If those 4,500 homes are added to the extra 1,900 homes suggested by the NHDC study of Central Bedfordshire's own requirements within the Luton HMA, that brings a total of 6,400 homes. This can be met by Central Bedfordshire's commitment to providing 7,350 homes in addition to the 32,000 it needs for its own purposes and at the same time eliminating the need for NHDC to build 1,950 homes on Green Belt land around Cockernhoe. However, the question once again arises which local authority is best placed to meet this unmet need. We contend strongly that Central Bedfordshire is far better placed to meet this need, especially as the need is contained within its own area of the Luton HMA, as stated by NHDC.
Our overall conclusions
In almost all of the scenarios examined as a result of studying in depth the NHDC paper (ED159) to the Inspector, the ONS 2018 housing projection statistics and its accompanying detailed datasets, as well as referral back to the NHDC Local Plan, the Luton Local Plan and the draft Central Bedfordshire Local Plan there is the overwhelming conclusion that there is likely in almost all scenarios to be no unmet housing need arising from within Luton.

There is a possibility that in one scenario, a small number of extra homes might need to be built to meet a residual unmet need arising from within Luton - at the most this looks to be around 500 homes. In this case the question arises where will these houses come from? Should it be from Central Bedfordshire - which is the second biggest component of the Luton HMA and with more identified building sites available, some not in the Green Belt, and with closer and easier transport links, services links and closer to the Luton jobs market - or- should it come from North Hertfordshire through building on the Green Belt around Cockernhoe and in the process engulfing the three villages of Cockernhoe, Mangrove Green and Tea Green - an area which has now recently been re-defined as making an overall significant contribution to the purposes of the Green Belt?

Clearly, there is no 'exceptional circumstance' demonstrated here to justify building on the Cockernhoe Green Belt.

But there appear to be more serious issues at play. It is very difficult to predict future housing need and it is important that the most robust and transparent methodology is used and that consistency is applied. Sadly, the arguments presented in ED159 have demonstrated statistical gymnastics and a cynical and duplicitous approach adopted by NHDC.

Such is NHDC's apparent desperation to build these homes, that they have now changed the goalposts by suggesting that the homes are not - as everyone has consistently believed for many years - to meet unmet needs arising from within Luton itself but, they now argue, to meet needs arising from within the wider Luton HMA. This goes right against the explicit statements in the MoUs and Statements of Common Ground signed by NHDC, with Central Bedfordshire, Aylesbury Vale and Luton Borough Council which make direct reference to the housing need arising from within the Luton Borough Council authority area.

The figures to justify this simply do not add up. NHDC should produce robust and clear justification as to why they have calculated that extra homes will be needed in the Central Bedfordshire area of the Luton HMA under the new ONS statistics, when Central Bedfordshire itself states that it see these new statistics leading to an overall reduction in homes required. In any event, NHDC does not appear to have produced these figures with any sign of co-operation or verification from either Luton or Central Bedfordshire.

NHDC's conclusions to ED159 state that under the various alternative scenarios discussed, 'the unmet housing needs from Luton could decrease. Viewed in isolation, this could call into question the justification for the allocations proposed to the east of Luton in the current Plan. However, this issue needs to be considered in the context of the broader Luton HMA's ability to absorb the shortfall'.

It goes on to say: "Under the alternative scenarios, Central Bedfordshire could have to provide additional homes in order to meet its own requirements. As such, although the unmet needs from Luton are lower, Central Bedfordshire's potential ability to address them could be reduced to an equal or greater extent. Once the potential combined requirements for Luton and Central Bedfordshire are taken into account, the east of Luton sites would still be required to make a positive contribution towards housing needs from the wider housing market area under the alternate scenarios".

NHDC also must explain why it is so keen to continue in its quest to build 1,950 homes in the Cockernhoe area, despite there being scant evidence of any unmet need arising from Luton, and also having no regard to Luton's over-delivery of housing numbers over a 27-year period. Yet their own figures, using the Lichfield methodology, suggests there to be no more than a notional increase in housing from Luton and Central Bedfordshire and this representation demonstrates, in all probability, that these homes can be met from within the Luton HMA by the number of homes able to be built by both Luton and Central Bedfordshire.

In short, NHDC's argument that 'exceptional circumstances' for the use of the Green Belt land to the east of Luton remains applicable is utterly wrong and should be dismissed totally.

This area of land around Cockernhoe should be kept in the Green Belt and, furthermore, the firm offer made by NHDC to its neighbouring authorities in the Luton HMA to build 1,950 homes in this area should be scrapped, as clearly this need does not exist anymore. NHDC seemingly has bent over backwards under the Duty to Co-operate with neighbouring authorities in its desire to build these homes so it can be argued that it has met its obligations under the Localism Act.

Even further, the idea that the narrow strip of land to the east of Luton with its tiny and statistically insignificant population of 1,700 people (500 of which are in the Cockernhoe area) should be part of the Luton HMA (322,000 population) in the first place is a nonsense and a convenient contrivance that has allowed, either - the unintended consequence of this small area potentially having to bear an utterly disproportionate and destructive contribution to meeting neighbouring authorities' unmet housing needs, or, - a deliberately and knowingly-created situation to enable NHDC to build homes for reasons other than to meet other authorities' unmet housing needs through the Duty to Co-operate.

Finally, why has NHDC not embraced positively the possibility that there may not be an unmet housing need arising from Luton Borough Council and in so doing protect the Green Belt that its constituents/residents so value and wish to preserve?