Written evidence submitted by Centre for Cities [HLV 049]

 

The main land value capture mechanism used in England is Section 106 agreements. This mechanism now provides over 50 per cent of total affordable housing supply in England. Between 2018 and 2022, Section 106 delivered an average of 26,700 affordable houses per year. Its contribution is therefore substantial and has helped to offset cuts to grant funding for affordable housing.

However, Section 106 faces several limitations as a land value capture mechanism:

Firstly, what Section 106 can deliver is dependent on the rate, location and type of private housebuilding occurring in a given location. Potential land value capture is therefore ‘bounded’ by the privately initiated development pipeline which doesn’t necessarily align with where housing and other infrastructure is most needed.

Secondly, the process by which Section 106 delivers affordable housing is disconnected from the bodies responsible for buying and managing that housing (mostly housing associations) and those responsible for managing housing needs more generally (local authorities). Recently, this has resulted in perverse situations where developers cannot find buyers for the affordable housing they have delivered, slowing the rate of both private and affordable housebuilding[1].

Thirdly, there is good evidence that Section 106 agreements and related planning policies do not effectively suppress land prices in high value locations to maximise land value capture. As discussed below, Centre for Cities’ modelling suggests a public-led land assembly model would capture a greater proportion of land value uplifts following planning permission for the public benefit.

 

Delivering more affordable housing and public infrastructure depends on both how effective land value capture mechanisms are at securing ‘public benefits’ through the development process, and the overall rate of development. Improving land value capture when land values change following the granting of planning permission therefore rests on a) ensuring land is bought at lower prices ahead of development b) ensuring that land comes forward for development, despite it transacting at a lower price.

One direct way by which the Government could try to achieve both a) and b) is through public bodies - development corporations, combined or local authorities - playing a larger role in pro-actively assembling land and allocating it for development. Compulsory purchase powers (the ‘no scheme’ principle for large sites, and now, following LURA 2023, if the scheme is deemed in the ‘public interest’) enable public bodies to buy land at close to existing use value – either through actually using powers, or, more likely, using them as leverage. The public authority would subsequently rezone and prepare plots for development, securing the value uplift by selling these plots at market rate to housebuilders.

This approach can give greater certainty that land value capture is being maximised as it:

This approach can also help avoid issues with planning policies which ask for more than is viably deliverable, which can slow the rate of housebuilding.

Centre for Cities modelling[2] suggests:

Figure 2 compares actual and modelled affordable housing delivery by region[3]. For an explanation of how the model works, please see Appendix A.

The bars in green show modelled affordable housing delivery if a) Government targets were met by private housebuilding and b) land were acquired at 5 times agricultural price[4]. Light green shows the theoretical maximum – what could be delivered if all surpluses were used to fund affordable housing.

In reality, affordable housing delivery would be lower than this maximum as land value capture would be expected to fund other public benefits. Dark green shows what could be achieved if only half of the potential land value capture were used to fund affordable housing.

Figure 2: Actual and potential annual affordable housing delivery through land value capture, summed by region (local authorities with sufficient greenfield supply only).

A graph with different colored bars AI-generated content may be incorrect.

Source: MHCLG Table 1011; Centre for Cities modelling. See Appendix I in Restarting Housebuilding III: new towns and land value capture for all data sources. Note: Modelling is static and figures presented are for the first year only.

The key finding is that, in each of the regions in the Greater South East, modelling suggests that significant improvements to the efficacy of land value capture mechanisms could be achieved. This is because this is a) where land value uplifts following planning permission are largest and b) evidence from modelling suggests that a significant proportion of this uplift is not currently being captured.

Figure 2 shows that even if private housebuilding rates met Government targets (purple[5]; sum=18,400), affordable housing delivery would fall short of what would be possible if land value capture were maximised and all surplus used to fund affordable housing (light green; sum=53,600). Even if only half of the potential surpluses were used to fund affordable housing (dark green; sum=26,800), the model suggests that current land value capture mechanisms could be significantly improved upon.

This gap between what is theoretically possible and what is currently being achieved suggests that present land value capture mechanisms are not able to (or are not being used to) suppress land prices to near 5 times existing use value (the land price assumed in the model). This failure is the main justification for a greater role for public authorities in land assembly.

Elsewhere in England, the scope to deliver more through land value capture is limited, if possible at all. If housebuilding rates increased to meet Government targets and the present level of Section 106 affordable housing delivery were maintained (purple, sum = 16,000), more affordable housing would be delivered than the model predicts (light green, sum = 10,500)[6]. This is because, in lower value areas (across much of England outside of the Greater South East) market rate housing doesn’t sell for much more than it costs to build (excluding land costs), so land values don’t change much (if at all) when planning permission is granted. There is limited scope for land value capture, local landowners’ expectations about land prices are low, and current land value capture mechanisms appear to be doing better than the model predicts.

In reality, the geography of potential land value capture is more complex than presented here, and the figures presented should be taken as indicative. But this illustrates clearly that vastly different house prices lead to very different possibilities from land value capture, and the results of the modelling strongly indicate that existing land value capture mechanisms could be improved upon in high value locations.

As discussed above, the results of Centre for Cities modelling suggests that scope for improving land value capture in regions with lower average land values is limited, at least by comparison to improvements made in places with higher land values.

Increasing land value capture in these locations requires the market value of developments to be higher, or costs to be lower, than predicted in the model. This may be possible for some, smaller sites, but is unlikely to be possible across places with lower land values in general, and is especially unlikely on larger sites. It is possible that a large new town-type development that was of very high quality could be very desirable and average house prices would be higher than they have been in the recent past. But any large-scale development would also involve flooding the local housing market, which would likely reduce average new build house price and thereby erode potential land value capture. 

The focus for improving land value capture through land-assembly led mechanisms should therefore be in locations where potential land value capture is greatest.

 

The economic opportunities of pursuing policies to improve land value capture in England are primarily that more affordable housing and other public benefits could be provided at lower cost to the Treasury, in some locations, at least over the medium-to-long term.

Achieving higher land value capture through public-led land assembly depends on the capacity and experience of public authorities. While it is normal in other countries such as France, Denmark and the Netherlands, this is not the way most housing development has happened in the UK historically. Capacity will need to be built within local authorities and a larger number of development corporations. They will need to develop expertise in site identification and site assessments, masterplanning, land acquisition (including ability to carry out compulsory purchase orders, if necessary), project management, and tendering a wide range of contracts. Much of this expertise currently exists in the private sector. Some of it, especially for the largest new town projects, is not plentiful in the UK currently.

The Government should therefore focus its attention where developing this capacity would realise greatest possible returns. As explained above, this would be primarily in the Greater South East. New Towns and other urban extensions offer opportunities to build this capacity across discrete projects, but a public land-assembly and contracting-type model can and should be used in a wide range of settings to maximise land value capture.

 

Beyond capacity and expertise required to manage such projects, a key challenge for public authorities will be access to finance and its sequencing. The above model of development requires public authorities to be prepared to spend money on buying, planning and preparing plots before they see the return from selling them later. For larger projects with significant infrastructure requirements, these costs can be substantial.

One option available to development corporations or public authorities would be borrowing on capital markets[7]. The least risky projects (those in higher value locations) could secure lower rates than available through the Public Works Loans Board. This is the approach taken regularly for funding infrastructure and urban development projects in Europe, including currently for the Grand Paris Express, a large metro expansion and housing development project around the French capital. If a project can secure over 50 per cent of its financing from the private sector, it can also be considered off balance sheet for net public debt financing purposes[8].

 

Besides the considerable challenge of building public-sector capacity, another potential challenge is the impact a larger role for public authorities in assembling land could have on the private housebuilding pipeline. The ability of public bodies to acquire land at close to existing use value presents a risk to private developers who have acquired (or intend to acquire) the land at a higher price, as they may lose out if a public authority acquires the land before they develop it. Forthcoming changes to CPO legislation which make it more straightforward for public authorities to set aside ‘hope value’, increases the range of scenarios in which potential private investors may be concerned about this risk. The Government and planning authorities should actively minimise this risk through:

 

Policies and strategies used to capture land value uplifts vary around the world, according to land market conditions, institutional capacities and legal frameworks. Policies which maximise theoretical land value capture are those in which there is public authority over land prices or land readjustment processes.

Land value capture following public land assembly has been effectively used in a range of European countries, including the Netherlands, Denmark, and France.

For example, Montpellier has been one of France’s fastest growing cities for multiple decades. It has achieved housebuilding rates well over 2 per cent per year while managing a rapid reduction in car use with a growing and thriving tram network.

Key to the functioning of this system has been Mayoral-led spatial planning, developed through the 1990s and finalised in 2001. This involves a cross-commune 15-year spatial development plan covering housing and transport infrastructure, stipulating expected densities and growth areas.

At the centre of the development process has been a multi-decade programme of public land assembly and a development company co-owned by the municipality and the state investment bank, called SERM. SERM plays multiple roles but is mostly focussed on coordinating and promoting development in a series of priority development areas or ZACs (Zones d’ Amenagement Concerte) located inside and on the outskirts of the city. Where land isn’t already owned, the municipality has the power to acquire it within these zones, and it has built up a further twenty years of land reserves in addition to existing ZACs.

A key advantage of public-led land assembly is that SERM has been able to generate competition between housebuilders for pre-determined delivery contracts. Housebuilders compete for secure, low-risk opportunities, and they do so by building high-quality, architecturally interesting buildings for a relatively low return. The uplift in land values also helps support high proportions of public housebuilding. While across France, there is a requirement to build 20 per cent social housing, these zones achieve even more – one third market sale, one third sold at a discount with repayment requirements when sold, and one third social rented.

SERM operates in a comparable way to UK development corporations, or authority-led regeneration vehicles such as Barking & Dagenham’s Be First, but has done so over a larger geography and for a longer period of time. Mayoral development corporations could be the appropriate vehicle to emulate the Montpellier model in the UK.

 

Improving land value capture does not require major legislative changes. The main challenges to improving land value capture in the UK are the capacity, expertise and pro-activity of public authorities to act in and shape the land market, as discussed above. 

The tools for maximising land value capture already exist, especially following changes to compulsory purchase legislation and the establishment of development corporations in LURA 2023. Proposed changes in the Planning & Infrastructure Bill represent minor improvements which should help make processes more straightforward, with less being decided centrally by the Secretary of State.

The development of public capacity to act directly in the land market should be expected to proceed alongside continued use of Section 106 to capture land value uplifts on privately initiated developments. The policies which guide Section 106 should also continue to evolve, responding to changes in costs and market value of developments, including lowering expectations on developers where it is clear changes to costs render development non-viable. In the longer-term, more widespread use of a public contracting model utilising public land assembly should reduce the use of Section 106 agreements.

See above regarding what is required to realise the potential of public land assembly as a mechanism for land value capture.

 

The delivery of New Towns offers an opportunity to develop capacity and expertise in public land-assembly led mechanisms for land value capture. As discussed in Centre for Cities’ December 2024 report, if these are focussed in locations with higher house prices, the potential for land value capture to fund a significant proportion of the costs of the projects, including high proportions of affordable housing, is clear (not withstanding idiosyncrasies between sites selected and above discussed challenges with sequencing of financing).

The rationale for pursuing a public-led model, including land assembly, for New Towns (and other large projects that fall outside the scope of the new towns project) is greater than it is for smaller projects because these types of projects are more complex and have longer time horizons. They are therefore perceived to be more risky and likely see returns from initial investment more slowly than smaller projects, so are less appealing for private investors. A public-led model for these large projects (likely led by a development corporation) therefore both increase the chance of them happening and the chance of maximising land value capture.

 

Support: pursuing more widespread public land assembly should help increase both the supply of land for housing in strategic locations and help secure maximum land value capture on those developments. The extent to which this contributes to the overall target will depend on whether the locations chosen are the right ones and the scale and number of projects pursued.

Evidence from Centre for Cities’ work suggests that the greatest opportunities to support housing growth and maximise land value capture are in the Greater South East.

 

 

 

 

Appendix A: Centre for Cities modelling

Centre for Cities has modelled potential land value capture across the country as part of research conducted for our December 2024 paper: Restarting Housebuilding III: new towns and land value capture. The results of the modelling (on potential new towns sites, rather than across whole local authorities) are presented in Centre for Cities’ report.

Centre for Cities would be happy to share and discuss this work with the HCLG Select Committee.

Discussion in this submission is limited to modelling of affordable housing delivery, to enable comparison with actually delivered through Section 106 agreements, but the modelling work can be used to understand potential land value capture more generally.

As illustrated in Figure 1, the model calculates an estimated cost of delivering a large-scale, greenfield housing development in each local authority. These costs include building houses, buying land, and delivering infrastructure such as roads and electricity, but does not include schools, GP surgeries, or any affordable housing (things which might be required in Section 106 agreements). The model also uses data on average new build sale price by local authority to estimate the potential market value of the houses built.

Figure 1: Cost and value inputs for modelling potential land value capture

A diagram of a project AI-generated content may be incorrect.

The difference between market value of houses built and total costs is taken to represent potential land value capture. The model is not prescriptive about how this potential land value capture might be realised, but it best reflects a public-led land assembly model where the price land is bought is, more or less, known.

An important variable in the model is the cost of land. The model assumes that land is bought at a fixed price, reflecting what would happen if either the land was assembled by a public body with compulsory purchase powers, or obligations placed on private developers worked effectively to suppress the land value. The results presented in this submission assume land is bought at 5 times agricultural price.

Modelling affordable housing:

The model assumes to build an affordable house, it is switched for a market value house on the same development – the total number of houses built (and therefore the costs of development) remains the same.

The cost of building an affordable house is the difference between the value of a market rate property and the rents received if it were an affordable house, as shown in Figure 2. Affordable housing in the model is 50% for social rent and 50% ‘affordable’ rent (80% of market value). Note, this is a higher proportion of houses for social rent than achieved in most affordable housing delivered by Section 106, and ‘affordable housing’ in the model is therefore more expensive to deliver than it has been in the recent past.

Where the difference between the social and market values are greater, the cost of building a affordable house is therefore also larger.

 

March 2025

[1] Eve, P & Kehoe, D (2024) The challenges of unlocking Section 106 delivery. London: Savills

[2] Based on the same methodology used for Centre for Cities recent report, adapted for use at local authority level: Lange, M (2024) Restarting Housebuilding II: New Towns and Land Value Capture. London: Centre for Cities.

[3] The model is only appropriate for greenfield development, so results presented are limited to local authorities where there is sufficient greenfield land to meet Government housebuilding targets for the next 20 years (hence, London only includes a small number of authorities) (N=228 out of 295).

[4] 5 times agricultural price is a higher price than could be achieved through compulsory purchase but is chosen to reflect the likely scenario that faster land transactions could be achieved by offering a price higher than the landowner knows they would get after more time-consuming legal proceedings.

[5] This assumes that the average value of developments in each authority would not decrease if housebuilding increased. In reality, it is likely that house prices would decrease with increased supply, meaning these figures are likely slight overestimates.

[6] This is partly because the model is relatively conservative, including costs incurred in large-scale developments that wouldn’t be necessary for smaller developments. Small developments with lower costs can enable larger surpluses than the model predicts, and therefore greater land value capture and more affordable housing delivery. It is also partly because affordable housing in the model is 50% for social rent and 50% ‘affordable’ rent (80% of market value). Most affordable housing delivered in the last 5 years has a lower proportion for social rent (if any) and therefore is cheaper to deliver than that in the model. Note however, both of these reasons are also true in the Greater South East, so while they explain some of the difference between actual and modelled affordable housing delivery, they doesn’t explain the difference between regions.

[7] Aubrey, T (2024) Avoiding the pitfalls of private finance initiatives and departmental budgets to fund the next wave of sustainable new towns and urban extensions. Cambridge: Bennet Institute for Public Policy.

[8] Office for National Statistics (2023) Monthly statistics on the public sector finances: a methodological guide.