Written evidence submitted by Shelter [FPS 154]

1.        About Shelter

1.1.  We welcome the opportunity to respond to this inquiry.


1.2.  Shelter helps millions of people every year struggling with bad housing or homelessness through our advice, support and legal services. And we campaign to make sure that, one day, no one will have to turn to us for help.


1.3.  We’re here so no one has to fight bad housing or homelessness on their own.

2.        Introduction

2.1.  Shelter is an unashamedly pro-housebuilding charity and believes that only by building more homes, and in particular more social homes, can our national housing emergency be addressed. An emergency that currently leaves hundreds of thousands of people in England homeless, millions trapped in unsuitable and expensive private rented housing and that continues to fail in delivering the homes we need as a country.


2.2.  Shelter is clear that the only way to solve the housing emergency that we face today is by dramatically increasing levels of social housing delivery and we support any move or call to make this a reality. Ultimately, we know that at least 90,000 new social homes are required each year to meet housing need.


2.3.  As an organisation we recognise that the planning system can, and should, be improved to deliver better results. Indeed, we have been at the fore in campaigning for this in recent years calling for improvements to Section 106, working with government to close the viability loophole and campaigning in local areas like the City of Manchester to improve delivery of social housing at a local level. We have also advocated for further improvements so that planning obligations get more social homes built, such as removing the small sites exemption and considering the extent of lost affordable housing contributions as a result of permitted development. So, we’re no strangers to the shortcomings of the current system or to thinking about how it could be better.


2.4.  In addition to this we support many of the ambitions government has stated it would like to achieve via planning reform. For example, supporting the idea of transparency in the developer contribution system and agreeing wholeheartedly that a greater share of planning gain must be captured for the public good.


2.5.  However, planning reform cannot alone solve the emergency that we currently face, and planning is not the biggest barrier to building more homes overall or more social housing in particular. This was the finding of the government’s own Independent Review of Build Out, led by Sir Oliver Letwin, which demonstrated that market absorption dictates the pace at which developers build.[1] It accounts for why increasing planning permissions doesn’t lead to a corresponding increase in housebuilding and won’t without other reform.[2] And it’s also recognised by the general public. In a recent large poll for us by YouGov, ‘wholesale planning reform’ was the least popular of ten housing priorities, with the support of less than 9% of respondents.[3]


2.6.  Shelter also has serious concerns about some of the reforms being proposed by government as part of its current approach to reform of the planning system. We have set these concerns out in both this submission and, in more detail, in our response to the government consultation on its planning white paper. Put simply though, we currently fail to see how the reforms being proposed by government can ensure that more social housing will be delivered, and we believe they bring with them a high level of risk for the small amounts of delivery we currently achieve.


2.7.  In this response Shelter has focused on answering questions 1,2, 4, 7 and 8 from the list set out by the Committee.

3.        Question 1: Is the current planning system working as it should do? What changes might need to be made? Are the Government’s proposals the right approach?

3.1.  Shelter is a pro-development charity and we believe the planning system can deliver better outcomes for communities than it does currently. We want to see major changes to our housebuilding system and our focus is on supporting changes that will deliver the social homes that the country needs.


3.2.  We are not the first to say that in its scope and specificity the white paper is more like a green paper. There are too many proposals with too many possible tracks of implementation for us to take a comprehensive view on every proposal. So we’ve prioritised what we focus on by taking an initial view of the most immediately relevant part of the white paper for social housing delivery. In our view that is ‘Pillar 3’ and so our response to the government’s consultation and evidence for the committee focusses on the proposal to replace Section 106 and CIL with a consolidated infrastructure levy.


3.3.  We wholeheartedly believe that the share of planning gain going to genuinely affordable social rented housing should be maximised. And we support the ambition to improve on the tools currently available to do this, including those highlighted by the committee in their inquiry into land value capture.


3.4.  However, we are concerned that the new infrastructure levy will have to overcome some deep challenges if it is to succeed and we are concerned by the current lack of supporting evidence for the proposal. Although Section 106 can be improved, we shouldn’t underestimate the estimated £4.7 billion contribution it currently makes to affordable housing or downplay the risk of starting from scratch. In our evidence we recommend steps for how the levy should be developed and safeguards that should be imposed to mitigate this risk.


3.5.  Supported by case study research carried out by leading consultancy Arup (which we are appending to our response), we have identified three major technical challenges that the current proposals for a consolidated Infrastructure Levy must overcome:


  1. Setting a flat rate that can account for variation between developments.
  2. Ensuring essential infrastructure is delivered when it’s needed without putting social housing at the bottom of the priority list.
  3. Resolving the problems above without making the new system slow and cumbersome.


3.6.  Additionally, the white paper makes some proposals and leaves some unanswered questions that could critically undermine the delivery of onsite affordable homes if resolved unfavourably. For example, the white paper leaves open the possibility that developers may be able to choose to make cash contributions rather than in-kind affordable housing, which could lead to a reduction. It also proposes giving councils total discretion over how to use levy receipts.


3.7.  If the government is going to continue to develop the infrastructure levy, it must demonstrate proof of its concept before it proceeds to implementation. We set out four steps that should be taken to do this:


  1. Creating a robust, measurable and enforceable commitment on social and affordable housing delivery.
  2. Developing levy prototypes and testing on-paper, e.g. through computer modelling
  3. Testing the levy in the real world through a limited pilot.
  4. Rigorous consultation at every stage.


Key Questions for the HCLG committee’s inquiry

The infrastructure levy: three major challenges

3.8.  In the white paper and other public communications, the government has committed to deliver as much, if not more affordable housing through the planning system.[4] However, it has not provided any evidence for how this will or whether it can be achieved. Using research by leading consultant Arup, we have identified three major challenges that the levy must overcome if the government is going to be able to meet the stated ambition.

1. Setting a flat tax that can account for variation between developments

3.9.  The proposals outlined in Pillar 3 of Planning for the Future propose setting a flat levy above a minimum threshold of development value. The paper proposes that affordable housing could then be secured as an in-kind payment of the levy to secure on-site delivery.


3.10.  The challenge that the levy must overcome is accounting for the wide variation in the uplift in value that occurs when developments get planning permission. It is not clear that any flat rate levy and threshold, even set locally, can address differences in land values, land value uplift, remediation costs and build costs between sites.


3.11.  Thus, it’s unclear how a flat rate levy, even with a liability threshold, can avoid creating a system that is simultaneously:



3.12.  Case studies compiled by Arup illustrate major variations in viability and land value uplift across the country and between sites. Their work looks at a development in Elephant & Castle which had a very high relative development value, contrasting with developments in Erdington and Edgbaston in Birmingham, which are low value. Variations between sites are not only geographical differences between local housing markets. They are also determined by multiple site-specific characteristics. For example, Arup’s research included a greenfield development in Gloucester in a typically “medium-value” area but with low land remediation/preparation costs. As such there was high land value uplift and the planning authority able to secure significant affordable housing contributions. By contrast, a development in Leeds in a typically “high value area” but on a brownfield site with a high existing use value and significant land remediation/preparation costs, did not secure as high a level of contributions.


3.13.  There is no obvious way that a single levy with a threshold, even set at a local authority level, could deliver increased benefits on some of these case study sites without reducing them on others or making them unviable.

2. Delivering essential infrastructure at the right time

3.14.  Another central part of the levy proposal is that receipts will be collected upon occupation, so that the levy paid reflect actual sales values rather than anticipated value.


3.15.  Infrastructure is often required before constructing homes could begin or be completed, so some measure to access levy receipts before they are collected on occupation would be essential. The suggested approach is allowing councils could borrow against expected receipts to deliver infrastructure.


3.16.  Borrowing against expected levy receipts will be inherently uncertain. Development values (and therefore levy liabilities per home) will change between the development of a local plan, granting permission and the point at which developments are occupied (when levy receipts are collected). This places risk onto local authorities which in turn will limit the amount that already overstretched councils are able and willing to borrow to deliver infrastructure. This will impact local communities where infrastructure is not delivered alongside residential development, and slow overall build out where essential infrastructure is required before construction can begin, threatening the delivery of local plans.


3.17.  Putting aside the question of whether councils have the capacity or appetite to borrow significant new sums, whether they will want to take on the development risk of delivering new infrastructure, or whether there is moral hazard for developers in insulating them from this development risk, our big worry is that affordable housing will inevitably get pushed to the bottom of the queue. Where debt is used to pay for infrastructure upfront secured against future receipts, it will always be rational to pay it off before delivering any affordable homes. These costs can be very high. For example, Great Park in Newcastle included the equivalent of £3 million in road works.[5] Having paid out this sum, councils are likely to want this money back first.


3.18.  (Neither do these arguments only apply to infrastructure. Section 106 is also currently used to cover costs during a development and ongoing costs afterwards and the mechanism of borrowing to cover these costs also looks set to create pressure to push affordable housing down the priority list.)

3. Resolving the problems without making the new system slow and cumbersome

3.19.  It’s unclear that there are solutions to the challenges posed above that would also retain the anticipated benefits of the levy in terms of time and certainty.


3.20.  For example, requiring in-kind infrastructure contributions on sites that require them could remove the requirement for council borrowing. But this would be similar to Section 106 and take time to negotiate and agree. Similarly accounting for differences in land value and uplift between sites in a local area would require site level assessments or an appeal process so that desirable schemes that are made marginally unviable by the levy could go through a more extended negotiation process. Again, this would add complexity. As such, the supposed benefits of a flat rate infrastructure levy may be lost and the value of wholesale reform questionable.

Providing proof of concept

3.21.  No one doubts that Section 106 could be improved. Shelter has supported several key improvements that could be made to the existing system that could contribute to increased social and affordable housing supply. However, it’s also important to recognise the contribution that Section 106 currently makes – delivering half of all social housing completions in 2018/19, with affordable housing agreements equivalent to £4.7 billion. It would be foolhardy to put this contribution at risk by resolving to entirely replace this system before proof of concept for the infrastructure levy has been provided.


3.22.  To provide proof of concept and demonstrate the levy can increase social and affordable housing contributions, the following common sense four steps should be completed before determining to implement it.


  1. Creating a robust, measurable and enforceable commitment on social and affordable housing delivery – for the government’s commitment on affordable housing to be meaningful, it must set out more detail of how it will be measured and enforced (this is discussed in more detail below).


  1. Developing levy prototypes and testing them on-paper, e.g. through computer modelling – the government should demonstrate that the deep challenges posed earlier in our evidence can be overcome on paper. In particular, it should be able to compare prototypes for the levy with real world data on affordable housing obligations secured through Section 106 to show it’s possible to set a flat levy that leads to a higher share of planning gain being captured.


  1. Testing the levy in the real world through a limited pilot – no computer model can fully account for human behaviour. Therefore, once the government has a good degree of confidence that the levy will work, it should test it in the real world. Given that permitted development does not currently make contributions to affordable housing, it might make sense to start with these developments.


  1. Rigorous consultation at every stage – as the proposals for the levy become less vague and more specific, the government should consult again.

A measurable and enforceable commitment on social housing

3.23.  We remarked earlier that the government has committed to delivering at least as much affordable housing onsite through the levy as with Section 106. Given the huge need for social homes, we think it would only be worthwhile pursuing reform if it will lead to more social housing. In either case, the government must develop a much more detailed, measurable and enforceable commitment, if it is going to continue to develop proposals for the levy.


3.24.  Recent experience demonstrates the pitfalls a meaningful commitment on onsite social housing must avoid. A comparable pledge was made to replace every home sold through the Right to Buy when the scheme was revitalised in 2012. There was no clarity upfront about the technical details of the pledge or how it would be monitored. These only emerged over subsequent years. No evidence was ever presented to demonstrate the pledge was achievable. Subsequently, even after the technicalities became clear, the pledge was never met. Finally, as it was not backed up by any change in statute or enforcement mechanism, it’s unclear how the failure to meet the pledge would be put right or what its status is now.


3.25.  So, a commitment on social and affordable housing must include detail of:



4.        Question 2: In seeking to build 300,000 homes a year, is the greatest obstacle the planning system or the subsequent build-out of properties with permission?

4.1.  The number of planning permissions is not the main barrier to increasing the number of homes we build as a country or hitting the government’s target of 300,000 a year. Recent research by Shelter showed that 40% of homes granted planning permission 2011-2019 have not been built.[6] So, it is definitively not the case that simply increasing the number of permissions or the speed of permissions will lead to a corresponding increase in the number of homes being built across the country. The government will not get 300,000 homes built in England a year just by liberalising planning.


4.2.  This is particularly the case in the current economic climate. In the summer, research for Shelter by Savills suggested that the recession could lead to a significant drop in demand for new market homes to buy as unemployment increases and mortgage lenders become more cautious. This is a pattern that’s been seen in previous recessions. As a consequence, they suggested that without intervention we would see a big reduction in the number of homes being built over the next five years. Housebuilding in England is currently overwhelmingly focussed on building new homes for people to buy, and if people can’t afford to buy them then they don’t get built. Planning reform has no bearing on this and won’t help overcome it.[7] This is a pattern that’s been seen in previous recessions. As a consequence, they suggested that without intervention we would see a big reduction in the number of homes being built over the next five years. Housebuilding in England is currently overwhelmingly focussed on building new homes for people to buy, and if people can’t afford to buy them then they don’t get built. Planning reform has no bearing on this and won’t help overcome it.


4.3.  So, the biggest current obstacle to building more homes is the combined effect of having a housebuilding system overwhelmingly focussed on building homes to buy for market sale with an anticipated drop in demand. To support housebuilding in the immediate term – and keep its vital contribution to the economy and employment going as we hope to power out of the recession – we should be looking for ways to increase investment in social housing, which is not affected by changes in market demand in the same way. Increasing social housing investment is the orthodox policy response and it provided vital support to the industry in both the nineties recession and Global Financial Crisis.


4.4.  Over the longer term, the government’s independent review of build-out, conducted by Sir Oliver Letwin, found that the reason that large developers build slowly once they get planning permission is to deliver homes in-step with low market absorption rates. This means developers will only build homes at the rate that people want and are able to buy them without bringing down their price. Sir Oliver’s principal recommendation for speeding up build out and boosting the number of homes built every year was increasing the diversity of the types and tenures of homes that are built.[8] In practice that means Government increasing investment for social housing, especially in a downturn.


4.5.  Clearly, planning can have a bearing on tenure diversity. For example, if the proposed infrastructure levy led to a reduction in social and affordable housing delivered onsite, it would reduce tenure diversity. Likewise, the current proposals to increase the small sites exemption from affordable housing obligations will lead to a reduction in tenure diversity. However, planning isn’t the only determinant of tenure diversity and planning reform alone – even it’s not just focussed on increasing the number of permissions – won’t fix the problem of build out. Doing so will require both increased investment and other reforms including land reform such as proposed in the committee’s inquiry into land value capture.


4.6.  In their worst form the current proposals could even lead to slower housing delivery due to an overall reduction in onsite tenure diversity.

Key Questions for the HCLG committee’s inquiry


5.        Question 4: What approach should be used to determine the housing need and requirement of a local authority?

5.1.  We support having a standard method for assessing housing need, but this must explicitly include a measure of the need for social rented homes now and in the future.


5.2.  By focussing on sale prices and median incomes, the measure of affordability current proposed for the standard method will have little relevance for households who need social rented housing and people on low incomes. We therefore believe that a standard method must include:



5.3.  As noted above, it won’t be possible to deliver enough social housing to meet local need through the planning system alone, even if the infrastructure levy is a great success. However, detailed assessments of the need for social homes should remain free from assessments of the viability of delivering just using the planning system. Where the need identified cannot be met by developer contributions (Section 106 or the infrastructure levy), the detailed social housing need figures should act as a framework to guide the use of other levers available for the delivery of affordable housing, such as government grant for social housing.

Key Questions for the HCLG committee’s inquiry


6.        Question 7: What changes, if any, are needed to the green belt?

6.1.  Recent years have seen increasing numbers of parliamentarians and commentators seek to build consensus around a new approach to England’s – and especially London’s - Green Belt. Siobhain McDonagh MP has promoted the idea of removing Green Belt designation from land within one kilometre of stations with good access to Zone 1.[9] Likewise, Policy Exchange have suggested using currently-protected land in London and the South East to underpin the development of new, high-quality settlements affordable to millennials in and around London.[10]


6.2.  There may now be additional opportunities to promote a new approach to Green Belt release, if government can convince local communities that land will be used specifically to address the housing priorities that have come to the fore in recent months:



6.3.  In addition, it is vital that any change to green belt ensures that existing local communities’ benefit. In ensuring this there is a strong case to be made for a focus on social and affordable housing, which can play a vital role in ensuring the future of local communities.


6.4.  Green Belt land values have been relatively shielded from speculation, and much Green Belt in and

around London is in any case in local authority ownership. There are therefore opportunities to use

this land to meet development needs neglected by the mainstream housebuilding system, creating

genuinely additional housing supply. By defining the uses to which released Green Belt land can be

put before protections are lifted, government can harness the uplift in value attendant on planning

designation to build high-quality social homes at gentle densities.


6.5.  For example, HMG could specify minimum levels of social rent and other sub-market housing which

must be delivered on former Green Belt land through changes to planning rules, backed up by

capital grant for schemes which come forward and are deliverable within a suitably ambitious timeframe. Homes England could produce a model design code for these “Homes for Heroes” to guarantee good design and place-making. This could potentially form a compelling political offer, capable of gaining public support from campaigning organisations and of reassuring communities that Green Belt will be used only where it can deliver housing that is more affordable and better quality than could otherwise have been achieved.


Key Questions for the HCLG committee’s inquiry

7.        Question 8: What progress has been made since the Committee’s 2018 report on capturing land value and how might the proposals improve outcomes? What further steps might also be needed?

7.1.  Shelter welcomed the Committee’s 2018 report on capturing land value and supported the recommendations it made to government. In particular we support the policy of amending the 1961 Land Compensation Act to remove the concept of Hope Value, something Shelter has campaigned against for a number of years.


7.2.  Unfortunately, since the Committees recommendations were published, no progress has been made in ensuring a larger share of land value uplift is captured for the public good.


7.3.  Analysis from the Centre for Progressive Policy has previosuly shown how much could be at stake here, with amendment of the 1961 Act potentially unlocking an additional £214 billion over the next 20 years. [11] Currently though this uplift is flowing not to the community or the state but largely to landowners and developers as excess profit. Evidence from HDH Planning on behalf of the Country Land and Business Association suggests that landowners alone currently receive 35% of the total uplift, which is likely to be a conservative figure.[12] Similarly Paul Brocklehurst, the chief executive of the Land Promoters and Developers Federation has agreed that around 50% of the total uplift flows to landowners, agents and developers[13]. These figures tally with analysis from the CPP estimating that post-tax profit received by all intermediaries is 58% of the uplift in land value.[14]


7.4.  Ultimately what is clear is that much additional land value could be captured, and this sentiment is evidently shared by the government which is clear in the Planning for the future white paper that it is seeking to capture a bigger share of planning gain.


7.5.  However, depsite expressing a desire to achieve this the government has done little to pursue actions that would deliver it. This includes failing to act on both the recomendations of the Committee and those of the Independent Review of Build Out provided by Sir Oliver Letwin (although Shelter has raised concerns about the approach proposed by Sir Oliver).[15]


7.6.  The proposals to replace S106 and CIL with an Infrastructure Levy (examined elsewhere in this response) are the only steps government has taken in recent years to increase land value capture. However, it is wholly unclear how a new flat rate Levy will achieve this, or if it can.


7.7.  If the government is serious about land value capture it should look instead to the 1961 Land Compensation Act. Currently, this Act enshrines in law the right to ‘hope value’. Put simply ‘hope value’ is the value of land that may exist if the land were ultimately used for a purpose for which it currently has no permission. In other words, it requires landowners to be compensated based on what they might one day have received consent to do with their land.


7.8.  Removing the right to compensation for prospective planning consents from the Act would change the incentives on landowners and would send a clear market signal that land assembly is done on the basis of what is actually going to be delivered – rather than what might. This would enable a new generation of large-scale, strategic, developments to be planned and delivered, bypassing the need for much of the poorly-planned speculative development – including on the Green Belt – that currently generates so much resistance from communities.


7.9.  These new schemes will also be financed not through increased taxation but by capturing more of the billions of pounds of land value windfall that is currently accrued by a small number of private landowners.[16]


7.10.  Practically that would mean that where the intention is to deliver a mixed tenure scheme with high quality design, ample social housing and appropriate community infrastructure, the land could be assembled at a value which reflected that. In this scenario the landowner still receives a fair price – indeed their compensation or sales price would likely still be significant – but communities also see a real benefit from new development.



November 2020


[1] MHCLG, Rt Hon Sir Oliver Letwin, Independent Review of Build Out

[2] Shelter, 40% of homes granted planning permission go unbuilt, accessed on 28/10/2020

[3] Shelter, Building our way out, October 2020

[4], BBC News, Boris Johnson defends ‘long overdue’ planning overhaul in England, 06/08/2020.

[5] Arup, Planning for the Future White Paper: Testing the implications of the Infrastructure Levy through Case Studies, October 2020

[6] Inside Housing, ‘40% of homes granted planning permission not built, says Shelter’, 07/09/2020

[7] Shelter, Rescue, recovery and reform: housebuilding and the pandemic, June 2020

[8] Rt Hon Sir Oliver Letwin, Independent Review of Build Out: Final Report, October 2018

[9] City Monitor, Siobhain McDonagh MP: Why I’m leading a campaign to build a million new homes on parts of London’s green belt, May 2018

[10] Policy Exchange, Tomorrow’s Places. 2019

[11] Centre for Progressive Policy, Gathering the Windfall, September 2018

[12] Country Land and Business Association, Written evidence to the House of Commons Select Committee for Housing, Communities and Local Government, available here.

[13] Oral evidence to the House of Commons Select Committee for Housing, Communities and Local Government, available here.

[14] Centre for Progressive Policy, Gathering the Windfall, September 2018

[15] Shelter Blog, Letwin identifies the problem but misses the solution, October 2018.

[16] See Centre for Progressive Policy, Gathering the Windfall