Monday, 9 August 2010

KAPF Closing Statement to Bondway Inquiry

On behalf of

I sat last night, having measured my trees, at 7 pm in Vauxhall Park, amidst the joggers and the picnickers, the tennis players and footballers, the dog walkers and the departing children, the homebound commuters and shoppers, and watched the weak sunshine making its way through the tree canopy, and listened to the wind soughing through the trees, from time to time drowning out the muffled sound of train and traffic. And as an extra picnic party arrived at 7.30 pm, to occupy an area even then being swept, in the mind’s eye, by the Bondway Shadow, with my acetates of the proposed tower to hand, I asked myself what possible “material considerations” could justify plonking down this monster of a tower in our tranquil backyard. Let us see how the arguments stack up.

As I conceded in Opening [CD8/4], the people of Vauxhall now know, whether they like it or not, that they have been well and truly clustered, and Mr Philpot drew my attention to a relevant London Plan extract [CD4/1] in cross examination. The issue is not whether there is to be a cluster, but what sort? And courtesy of the Freedom of Information Act, we have procured Lambeth Council to disgorge two urban design studies of the issues and possible forms of cluster. There are four on offer, only one of which would countenance a building of the proposed Bondway Tower’s height (149m) on its site.
• VUDS 2006 [K1.14] - a clump round Vauxhall Cross/Triangle, with a peak height there of 140m
• 2008 Draft Vauxhall SPD [CD5/2] – a similar clump, but with no peak height prescribed
• 2009 Draft VNEB OAPF [CD4/3, hereafter “OAPF”] – the least particularised, the crenellated table top with gaps, with buildings “in the region of 150m”
• 2009 Design Study [K1.15] – a preferred rising profile towards the north

And both the planning documents acknowledge the need to temper the impact of tall buildings on heritage assets:
• “Strategic viewing corridors, river prospects, local views and the historic environment require detailed consideration in relation to any tall buildings in the Vauxhall Heart, and proposals which would adversely impact upon those elements will not be accepted.” [Draft Vauxhall SPD] [CD5/2]

• “there are a number of key issues to be considered in terms of the emerging tall buildings strategy for the OA: - ... The potential for tall buildings to compromise key local views from conservation areas both within and beyond the OA” [Draft OAPF] [CD4/3]

Consistently with this general position, three out of four of these studies reduce the height of the tall buildings in their cluster nearer the conservation areas to the east. Only the OAPF, with its unparticularised tall buildings blob at Vauxhall would allow the Bondway Tower at 149m to even get to the starting line.

The Bondway Tower would be the effective precedent, setting the form of the cluster for the foreseeable future, and we think these urban design considerations are highly material considerations against the present application.

There must also be a question mark whether a policy free developer-led tower building campaign will give us anything but residential towers alone, with no significant retail or leisure component. Will that actually regenerate Vauxhall – give it a heart and make it a “destination”– as the developer claims? It is notable that after 10 years of residential led regeneration in the form of St George’s Wharf, Tuckermans can comment on Vauxhall

“Vauxhall as a location is well provided for by transport facilities, benefiting from the
train and tube lines as well as the new bus station. It is also a short walk to Westminster and Victoria. However, retail amenity is poor and the local environment has no ‘heart’ from either a community or business perspective. It is also viewed as a harsh environment due to the one way system, the clubs and the vagrant population. It is therefore generally seen as a secondary office location by the property industry and business community.” [A7.3 para 2.4]

Throughout this Inquiry we have been exercised by the sheer density of the development, which in residential terms come in at around three times the upper limits of the Mayor’s density matrix [CD4/1], and in plot ratio terms around 20:1. As we noted in proof [K1.1] and evidence in chief [K1.17] these recent residential towers are way beyond the ranges contemplated by the density matrix, beyond even the 500 units a hectare upper limit of “Living at Superdensity” [CD6/8], referred to in proof.

In our view we are sleepwalking into a wholly different range of buildings, without assessing the impacts of living in these ways. We note that few if any of the recent appeal decisions for tall residential towers raised these issues, and the mix of dwelling sizes generates a “child lite” development on policy assumptions about child yield, suggesting unsustainable demography and unstable communities. So for example, the child yield from the proposed development is 117, out of an estimated 880 inhabitants, a child proportion of about 13%. This compares with a Lambeth proportion of 19% [2001 census] [CD6/8] and the 25% proportion embodied in the OAPF [CD4/3]. In addition, amenity space has to be contrived in unusual ways, (it proved impossible to provide open air child amenity space on this site, and the developer has to look to an existing park to satisfy this requirement) and there is a serious issue about pressure on public amenities and services from such a concentrated development.

For these reasons we regard the wholesale setting aside of the density matrix as a significant departure from plan (both London and Lambeth) which has not been justified. We agree strongly with the evidence of Lambeth’s witness Mr Rees that this is overdevelopment, and the proposed tower would be an overbearing development, contrary to UDP Plan Policy 33 [CD5/1] .

The employment arguments have been well trodden during this Inquiry, and we put them again for the record. The development site forms part of an MDO [MDO 81] within a Key Industrial and Business Area [hereafter “KIBA”] [[CD5/1].
• On such a site, UDP Policy 22 provides that “Here some residential and other non-employment uses are acceptable on appropriate parts of the site. The overall development, however, should be predominantly employment-based, incorporating the maximum feasible amount of employment development -...”
• The policy gives developers anxious to demonstrate their conformity to policy two safe harbours in the policy itself, and an overriding one in the rubric to MDO 81 itself
• In this case we are testing whether a development 90% residential, with a commercial GDV of 1.5% (as emerged in the evidence of Mr Billington), and which actually reduces by 40% the existing employment floor space on the site is actually “predominantly employment-based, incorporating the maximum feasible amount of employment development -...”
• As appears from the common rubric to MDOs 80 to 83, each MDO is a site, and some are also groups of sites. In the latter case, one third of floorspace on each constituent site has to be for employment use, unless it can be demonstrated that, through comprehensive development, this amount of employment floorspace can be achieved across the entire MDO site. The developer is unable to acquire the remaining small site in MDO 81, and certainly cannot demonstrate that the one-third target can be achieved.
• It follows that the development does not conform to UDP Policy 22.

As regards affordable Housing, Mr Billington’s evidence has been enlightening, and the disclosure of the full viability assessment and expert appraisal [A6.5] helpful, and will enable the Secretary of State to have a better view of the working of the viability testing machinery. This regularly abates the 50/40% affordability headline rates in policy to barely half those levels, even with developments of up-market flats, which might be expected to have the capacity to deliver nearer policy headline rates.

The viability test essentially tots up the likely value of market sales and the capitalised value of commercial contributions, and deducts the likely construction costs, prescribed levels of developer profit and financing charges and any social housing grant, to assess how much headroom is left for affordable housing, while still leaving a residual value just above the existing use value [hereafter “EUV”] of the site. Crucial to this process is the estimation of flat prices in future years – in the case of Bondway, if it were to secure permission, the first estimate of prices will have taken place about 6 years before completion and actual sales (2009 assessment and a four year build completing in 2015 [see Construction Statement at CD2/4A para 1.44]). Such estimates are notoriously error prone – Property Forecasts, the leading forecaster in the field for town and city specific forecasts finds that even three years ahead, only 25% of its forecasts are within 15% of the true level.

What has emerged from witness testimony is how sensitive the % provision of affordable housing is to the estimates of market sales. Mr Billington testified that an increase of 5% in market sales estimates (and this was the difference between the professional views of those compiling the original viability assessment and of the Council’s expert appraisers of it) increased the affordable homes percentage emerging from the viability test from 23% to 32%, an increase of no less than 40% (9 percentage points) in provision.

We ourselves have doubts about the market sales assumptions currently embodied in the viability assessment, (around the £600 to £635 level per sq m), given current prices at the nearby St George’s Wharf development, anecdotally in the £700 per sq m range. This is only partly cured by the provisions for second and subsequent review, just before any permission were implemented, as, given the four year build time, this still involves estimation of flat prices four years ahead. We think such estimates are inherently uncertain, and likely to be conservative, and we think the evenhandedness of the methodology is in question.

The whole area needs a dose of sunshine, so that society can see whether the balance is being properly struck between developer and wider public interests, and to that end we consider that in this and similar cases, if permission were granted there should be a post mortem following completion, to audit the assumptions and estimates made, either as a stand alone audit by the Council, with developer co-operation, or via an addition to Sch 4 of the S106 agreement [CD1/10B], stipulating such a report by the developer.

English Heritage [EH] [CD2/19], Westminster Council, our Waterloo colleagues [W2.2] and Miss Hoey, the local MP have all expressed their concerns about the impact of this proposal on distant views. There is a risk of piecemeal erosion, and the impact of the proposed tower on the M12 view from the Cenotaph [CD2/4E], as noted by Miss Hoey, is notable. So also is the beginning straggle of tall towers in the distant view from Westminster Bridge [CD2/4E image M19 Cumulative], where either the cluster profile of the 2006 VUDS [K1.14] or the argued rising profile of the 2009 study [K1.15] are to be preferred on aesthetic grounds. We commend these opinions to the Inquiry and the Secretary of State and adopt the EH view as expressed in its representation:

“Our view is that we support good quality contextual modern architecture, which can enhance the historic environment given careful consideration of the relationship with its surroundings. By virtue of its scale and massing of the proposal, its significant harm on the setting and significance of the World Heritage Site, listed buildings and conservation areas, and loss of local heritage asset it is difficult to see how this could be achieved without a significant reduction in scale and height of the building.” [CD2/19]

As we say in our supplementary note on wide angle lenses [k1.16], which Mr Coleman uses for his 72 studies of the Tower [CD2/4E], they mislead the eye by providing extra foreground, reducing the significance of distant objects, like the proposed tower. For that reason, we regard Mr Coleman’s pictures as markedly understating the impact of the development, which becomes even more obtrusive and overbearing, particularly in its impact, broadside on, on Vauxhall Park and its conservation area. All the parties to the appeal agree that the only real assessment of the visual impact of the proposed scheme is to go and see the views personally, armed with acetates showing the outline of the proposed development above the existing skyline.

Given this circumstance, when the issue comes to rest with the Secretary of State, it will be necessary for the ultimate decision taker to be able to say, and record in their decision, that they themselves have looked personally at the views concerned before reaching their decision, and not merely relied on the misleading photographs in the case, else they will not have had regard to a relevant consideration.

The testimony of all the residents and community representatives who gave evidence to the Inquiry [eg R1 – R8 and evidence of Miss Hilling] was that the proposed tower was overbearing and intrusive to their community, their conservation area and the park it embraced. And they were already a vibrant, largely self built community, in no need of “repopulation” as the developer suggests.

As regards plan policies governing the impact of this broad tall tower on heritage assets, they are plentiful. We cite in particular:
• UDP Policy 33 – overbearing and developments affecting conservation area settings (overdevelopment issues cited above)
• UDP Policy 40 – tall buildings harming the character of conservation areas and impairing the development possibilities of surrounding sites without a wider masterplan (impact on Vauxhall and St Mark’s conservation areas, including Vauxhall Park in particular, and representations from neighbouring property about impaired development possibilities – and absence of master plan!)
• UDP Policy 45 – development adversely affecting the setting of a listed building (Woodstock Court)
• UDP Policy 47 – Development outside conservation areas harming settings and views in and out (Vauxhall and St Mark’s conservation areas above) [CD5/1].

We also remind ourselves of the terms of the draft SPD [CD5/2] and OAPF [CD4/3] cited above, which both recognise, in terms, the need to temper the design of tall towers at Vauxhall to take account of the impact on heritage assets.

In the circumstances, we think the case is made out that the proposed development breaches all the plan policies cited, and should be rejected on those grounds too.

Mr Philpot, in cross examination, sought to persuade me that, as regards a residential led development such as this one, there was no practical difference between the “employment led” areas washed pink in the OAPF land use diagram of OAPF p27 [CD4/3], and the “residential led” areas washed yellow. If he is right, there is little point in planning for the OAPF – it will be a developer free for all, with social infrastructure grudgingly provided by developers, if at all, and a question mark over whether it is capable of delivering the employment aspirations of the framework. We remain of the view that there is such a distinction, and wait to see if it is strengthened in subsequent drafts of the OAPF.

Both we and Miss Hoey, with her wide experience of the problems of Vauxhall, support the provision of a Metropolitan Police Neighbourhood office in the development, which is provided for in paragraph H and Sch 3 para 5 of the S106 agreement [CD1/10B], but only if the Secretary of State considers that it satisfies the now statutory tests of necessity, directness and proportionality. We argue that it does in broad terms – we draw attention in proof [K1.1 and K1.8] to the gap in police station presence in Vauxhall, the development adds 880 residents to an area with high levels of crime for robbery and serious violence to the person, and the 90 sq m of office space is hardly disproportionate to the overall scale of development, at upwards of 50,000 sq m. We urge the Secretary of State, were he to permit this development, to agree also this modest contribution to the public safety of its residents and their neighbours.

We drew attention in proof [K1.1 and evidence in chief [K1.17] to the issue of public transport accessibility, judged by PTAL rating, versus capacity of the infrastructure itself. In policy we now find recognition that both are important to the satisfactory location of development, and we note in particular UDP Policy 9 [CD5/1 and its London Plan equivalent] and OAPF p76 [CD4/3].

Transport assessments that concentrate on Underground train capacity at Vauxhall miss the point (at least at present) – what is at issue for the moment is gateline and escalator capacity, and we are near to various trigger points in passenger numbers that will increase periodic gate and station closures to deal with congestion in these access arrangements. We illustrated the consequence of such closures on successive days in proof [K1.6]. This Bondway development did not present a full cumulative analysis of transport impact [see CD2/4D], lacking material for recently consented developments like the Vauxhall Sky Gardens. The Secretary of State should be informed that subsequent developments, like the Vauxhall Triangle development, will need a more complete analysis, and when this is available, we may find that congestion begins to trump accessibility, unless there is significant expenditure on mitigating infrastructure.

As regards mitigating open space deficiency, we argued this in detail in proof [K1.1] and evidence in chief [K1.17]. It is clear that additional open space at Vauxhall is critical to the success of any cluster there:
“Within the Vauxhall Heart, intensification is expected to occur with new development creating a cluster of taller buildings of exemplar architectural quality. New buildings will be expected to form a cluster framing a central public space with building heights varying according to the relationship with the river and residential hinterland.” [Vauxhall draft SPD] [CD5/2]

“We support the 40m wide public space to the west of the site, which is proposed as part of Lambeth’s Vauxhall Area SPD, and feel that the success of the cluster is, in fact, dependant on the implementation of this. However, the ground floor of this building will need to work hard to animate the space, which will be of considerable scale. We urge the local authority to ensure that the public space is well designed, carefully integrated with adjacent new buildings and offers significant amenity to local residents.” [CABE Report on Bondway Tower] [CD2/18]

And we heard a similar view from Mr Rees in evidence, of the vitalizing and humanizing effect of well designed public open space on tall building clusters.

There is a clear audit trail in policy too:
• “50(c) New Open Space, Greening, and Green Chains – The creation of new open spaces, urban “greening” initiatives, and the linking and improvement of open spaces will be supported, especially in areas deficient in these features. Developments which materially add to the demand for open space, which are proposed in an area of open space deficiency, or where existing open space needs improvement, will be required to contribute to appropriate improvements in open space provision in the immediate area. Where on-site provision or provision in the immediate area is impractical or insufficient, developers will be required to contribute to such initiatives elsewhere. Arrangements for the longterm maintenance of new and improved open spaces will be secured.” [UDP Policy 50(c)] [CD5/1]
• The development lies in an Open Space Deficiency Area (see UDP Map 2, p135)
• There is a two part tariff for S106 contributions in Vauxhall and Waterloo, under the adopted SPD on S106 obligations [CD5/7]:
“Lambeth Open Space Deficiencies. The National Playing Field Association have a general standard of 2.4 hectares of open space per 1,000 population, Lambeth as a whole is deficient in meeting that standard, in that it has 1.54 hectares per 1,000 of population, projected to fall to 1.44. Lambeth’s Open Space Strategy recommends that a target of 1.6 hectares per 1,000 population is set for 2016.
2. New open space and investment into existing Parks. The creation of new open space will be encouraged on site or in the immediate vicinity. If this is not possible a financial contribution will be sought to increase open space provision in areas identified in the Lambeth Open Space Strategy, and/or expected towards investment in existing parks and open spaces.
3 Enhanced contributions Waterloo and Vauxhall – Additional higher financial contributions will be sought on a park or open space project basis from commercial and residential developments in these two opportunity areas to support environmental and physical regeneration objectives.”

And we understand that such enhanced contributions have been made in Waterloo, with Shell agreeing £1m for open space on a 37,000 sq m commercial development, and Elizabeth House agreed £1.5m for open space on a 280 unit development [CD7/4].

It is not for us to speculate here why the Council’s officers have failed to implement the Council’s adopted policies for open space contributions at Vauxhall. But it is clear that there is a need for new open space to vitalise the proposed cluster, and established machinery for securing contributions to finance it.

The 880 people who would occupy the Bondway tower generate an additional demand for 1.4 hectares of open space, at Lambeth’s 1.6 ha per thousand standard, and that in an open space deficiency area. Bondway’s pavements and back passage do not sensibly minister to this need. The generated deficiency is actually equivalent to half of Vauxhall Park (2.82 ha) [see CD6/16]. It is no use the developer piously hoping someone else will provide it – it is part of the necessary mitigation from choosing to develop a high density residential tower, and he should put up the funds to acquire mitigating space, or choose to develop something else that does not put excessive strain on the infrastructure.

I was chided in cross-examination for leaving the developer in uncertainty about the value of S106 contribution we would seek. Plainly we can’t have a developer left in uncertainty, and I calculate that 1.4 ha at EUV is £38m, and I would regard that as an adequate level of mitigation of this particular detriment of the development. However, given the terms of the SPD and CABE comments about open space at the heart of the cluster, we would be able to welcome £10m (0.4 ha, the size of the open space at the heart of the cluster, contemplated by the draft SPD [CD5/2], valued at EUV) as useful funding and a clear recognition of need.

This is a conservative measure of the damage that would need to be mitigated if the development went ahead. If you build what you are pleased to call vertical villages, you need to remember to provide the village green – if you build Georgian terraces stacked into the sky, you should not forget that ornament of urban living, the Georgian garden square. If we are told that it is quite impossible to fund a development like this and provide the necessary open space, we should not weep, and might invite the developer to try his hand at a mix of uses that does not look to his neighbouring developers to provide necessary infrastructure he does not care to, and over burden the existing public realm.

I have no doubt Mr Lindblom will talk of the benefits of park and playspace and public realm contributions, especially in an open space deficiency area – and even invite us to be grateful. But these are £1m solutions to £10m problems, and problems caused by the developer itself to boot.

My conclusion, sir, is simple. In this case you are creating the effective precedent for the Vauxhall cluster of tall towers. This is the:
• Wrong starting point for the Vauxhall cluster
• With the wrong mix of uses
• In the wrong form of tower
• With the wrong mitigation, and

I invite you to recommend the Secretary of State to reject the application.

David Boardman
Kennington Association Planning Forum

6 August 2010

Lambeth Core Strategy - Examination in Public - KAPF representations

Date: Monday, 2 August, 2010, 11:47

Simon Osborn, Esq
Programme Officer
Lambeth Core Strategy Examination in Public
Dear Sir
I attach further statements on behalf of the Kennington Association, in relation to
  • affordable housing
  • employment land
  • tall buildings
We are disturbed that no community association has been listed in your provisional list of participants, only developers and the Council, and that such an abbreviated timescale is being adopted, to deal with so many issues (4 days only, cf 8 days for the equivalent proceeding in Southwark).
We ask that we be listed for all four days' discussions
Yours sincerely
David Boardman
Kennington Association Planning Forum

Note by the Kennington Association Planning Forum

Lambeth Core Strategy
Examination in Public –Affordable Housing - Policy S2(c)

1 This Association is a Rule 6 party at the current Bondway/Octave Tower Inquiry, at which the full Affordable Housing Assessment and Independent Appraisal for the proposed, predominantly residential, development were put in evidence. From these, and evidence emerging at the Inquiry, it is possible to form a more detailed view of the methodology of viability testing using the “Three Dragons” toolkit.

2 The current issue of the toolkit requires, before the application is lodged, a comparison of likely costs of development, particularly construction costs, financing charges and developer profit at standard rates, versus likely proceeds, particularly the estimated value of ultimate market sales, and capitalised value of any rental element like office floorspace. The amount of affordable housing is computed as that proportion of the residential build, with or without housing grant, that is calculated to allow the residual value (proceeds minus costs) just to exceed the existing use value of the site. In the Bondway case, the affordable provision emerged from this process as 23% affordable with grant (cf the 50% headline rate in policy)

3 This calculation is performed well before determination of the application (in the case of Bondway in mid 2009, with appeal not likely to be determined until the last quarter of 2010). Typically it is subject to independent technical appraisal by outside experts on behalf of the Council, and in the Bondway case they estimated future flat prices as 5% more than the developer. It emerged in evidence that this higher figure would support not 23% but 32%, a full nine percentage points higher, a highly leveraged result.

4 This affordable provision is subject to a review mechanism, under which the process is repeated 3 months before implementation. This still involves the estimation of flat prices some four years before completion, which is a highly inaccurate process. As the attached note shows, the professional forecaster Property Forecasts, forecasting house prices only three years ahead for towns and cities, finds that only 25% of its forecasts are within 15% of the actual results (2009 actual compared with 2006 forecasts).

5 Given the sensitivity to estimated market sales, there can be no confidence that the methodology is delivering an accurate assessment of the capacity of developments to sustain affordable housing, and an inevitable suspicion that estimates will be made conservatively, to the detriment of affordable housing provision. There is no regular post implementation review of the assumptions made, once actual market sales proceeds are known, and if there is to be any confidence in the methodology, there should be. We think Policy S2(c) should be amended to provide for just such a post implementation review.

David Boardman

Kennington Association Planning Forum
2 August 2010

Note by the Kennington Association Planning Forum

Lambeth Core Strategy
Examination in Public –KIBAs - Policy S3(b)

1 This Association is a Rule 6 party at the current Bondway/Octave Tower Inquiry, at which a development with
·         90% residential floor space,
·         one and a half percent commercial Gross Development Value,
·         on a KIBA (MDO81 under the current UDP) requiring 30% employment floor space for any redevelopment
·         a reduction of 40% in employment floor space compared with the existing use of the site
argued that it should be allowed to proceed.

2 In the light of this development, pressure to bias “mixed use” development very heavily to the residential end, is likely to dramatically reduce employment floor space in aggregate. This would be to the detriment of employment uses, along with the de-designation of the Vauxhall KIBAs and the softening of the employment space test, so eg that provision of employment floor space of no more than 50% of site area would satisfy the policy. We oppose the de-designation of the Vauxhall KIBA, and argue that Policy S3 should have as an underpinning minimum, that no redevelopment on a KIBA should reduce existing employment floor space on that site.

David Boardman
Kennington Association Planning Forum
2 August 2010
Note by the Kennington Association Planning Forum

Lambeth Core Strategy
Examination in Public –Vauxhall, Tall Towers and Open Space – Policies S5(b), S9(d), and PN2(f)

1 This Association is a Rule 6 party at the current Bondway/Octave Tower Inquiry, at which the basis for a cluster of tall buildings for Vauxhall was examined, together with the issue of provision of additional open space, as an element critical to the success of any cluster. Through Freedom of Information requests, this Association was able to illuminate these issues by putting before the Inquiry two urban design studies of potential clusters at Vauxhall, commissioned by Lambeth Council, but previously unpublished, and these are attached to this note.

2 At least 4 sorts of cluster are on offer at Vauxhall, all in studies or draft planning documents, and nothing useful in adopted policy. As the 2006 urban design study foresaw, the absence of adopted planning guidance has meant that tall tower proposals are coming forward piecemeal, and some, such as Bondway, cite the Draft VNEB Opportunity Area Planning Framework’s suggested cluster as justification for a 150m tower to the south, close to conservation areas and Vauxhall Park, rather than the draft SPD’s proposed cluster with its roof lines rising toward the Vauxhall Cross site in the middle.

3 Supplementary planning guidance is urgent, and the cited policies are deficient, in that they fail to require new open space at the heart of any cluster. This was a significant element of the draft SPD (at p72) and was seen as critical by CABE to the success of the entire cluster, when reviewing the Bondway design. But developers are declining to provide it, looking to the next set of developers to do what they are unwilling to do themselves.

4 We strongly believe that these policies, particularly S9(d) and PN2(f) should be amended to require any cluster at Vauxhall (which is indeed in an Open Space Deficiency Area) to be designed round additional new open space, and that developers should expect to contribute to its establishment in kind or cash equivalent, as already provided for by the current Lambeth SPD for S106 contributions for open space in Vauxhall (Appendix D1).

David Boardman

Kennington Association Planning Forum 
2 August 2010



Historic accuracy of 3 year forward flat price estimates – Note By KAPF

Source: – 29 July 2010

Historic accuracy figures summary, Towns & Cities (tested Sept 2009)

As with the Neighbourhoods, comparisons are between forecasts we made in Aug 2006, from Land Registry data to June 2006, and filtered Land Registry data up to the end of second quarter 2009.  Averages for Towns & Cities are calculated from postcode sector (neighbourhood) forecasts, using (mainly) the Royal Mail definitions for towns and averaging relevant sector values .
·         Overall for houses, around 81-86% of three-year historic forecasts are classified A or B, down around 10% from last year;  25-29% are classified as A, significantly down from last year’s 64-70%.
·         Around 38 % of A forecasts for houses are within 15% of the averaged filtered Land Registry values at June 2009, well down on last year, with only around 7% within 5%.
·         For houses, 27-28% of A and B forecasts are within 15%, and 25-26% of all forecasts (including U) are within 15% at the end of three years. For flats & maisonettes, this figure has fallen somewhat to 23% of A and B forecasts within 15%, and 20% of all forecasts.
·         Less than 2% of all house price forecasts are classified as U;  for flats & maisonettes, around 5.3% are classified U.
·         Around 26% of all forecasts are in the same overall direction as the smoothed Land Registry data. For A category forecasts, this is 33-37%.
·         Average error at the end of three years for A classified forecasts is around +16% (standard deviation SD of 6.5%), while that for A and B classifications together stands at around 21% (SD around 11.4%). For flats & maisonettes, average error for ‘A’ and ‘B’ classifications was +22%, with an SD of 11.7%.
·         Overall average error was between +22.8 and +23.7% for houses (SD 15-16%), and +28% for flats & maisonettes, with a somewhat higher SD of around 20%.
Once again , a ‘-’ sign against an error value would show an underestimate, while positive values show over-estimates.  All values indicate overestimates, as one would expect in the boom-and-bust climate prevailing over the last three years.
Flats & Maisonettes:
Proportion of dataset (%)
Average error
Standard Deviation of error
Proportion positively correlated
Average correlation
Within 5% after 3 years
Within 10% after 3 years
Within 15% after 3 years

David Boardman
29 July 2010