Thursday 1 October 2020

Planning for the Future: some solutions for our housing crisis at last?

By Paul Cheshire

The white paper, Planning for the future, published in early August 2020 represents the first serious attempt to reform our dysfunctional land use planning system since its inception. Although the Barker report’s diagnosis of how our planning system caused our housing shortage were well researched, the recommendations for change did no more than paint some steely surfaces over the crumbling fabric of planning. They were fixes for an unfixable system. This white paper promises more.

Its recommendations tackle two of the four policy-created barriers to building enough houses of the type, and in the places, people want to live in (adding extra uncertainty and risk to development; restricting land to build on; just saying you can’t; and giving communities no incentive to accept development: see my article for NIESR). If implemented cleanly and transparently the proposals would greatly reduce the risks and uncertainty our present system imposes on builders, while providing some incentive to local communities to accept new development and funds for the infrastructure that new developments require. At the same time, rather than destroying planning and undermining democracy as claimed by its critics, the proposals  would strengthen planning and retain full democratic local control ensuring standards of environmental protection were maintained while improving the quality of new development.

Central to our planning system, since its creation in 1947, is its discretionary and piecemeal decision making. This is really the antithesis of “planning”: every decision about significant development is a one-off made by local politicians. Decisions do not necessarily reflect the “plan”. This is not just because only about half of local authorities (LAs) have a valid local plan: even when they do have a plan, planning committees’ decisions often do not follow it.

Moreover, since there is no standard set of definitions nor format for storing or accessing local plans, it requires detailed local knowledge to understand what “the plan” says. This means it is also impossible to evaluate the system as a whole since there is no comprehensive database on plans (where there are plans). Researchers at LSE and Knight Frank spent a total of 18 months trying to generate comparable data on how much land LAs were allocating for housing. Even this effort was only able to produce data for 107 LAs and careful inspection showed that data from a total of only 73 were credible and could be adjusted so they were comparable.

So government may have elaborate, even sensible guidelines, as to how much land LAs should be allocating for housing but there is absolutely no way of evaluating the effect of such guidelines. The oldest of these guidelines is that LAs should supply land to accommodate expected population growth. This may be a very bad principle since it gives no weight to local affordability but the evidence from the 73 LAs – as far as it went – was that LAs did the opposite. On average the faster the local population had grown in the previous five years, the less land was supplied

So this gives us two reasons why our planning system adds risk and uncertainty to the development process: we do not know what “planning” is doing and we cannot anticipate whether any particular development proposal will be permitted. The white paper proposes to:

  • require LAs to make plans in a standard format, using a common GIS template (making the results transparent): and
  • move away from the current political decision on every significant proposal – effectively randomised decision-making – to a rule-based system. This would mean any new development consistent with the local plan and local design rules would almost automatically go ahead. As is the case in most Continental European countries now.

LAs would have to have plans which identified land according to three categories: i) Growth Areas where outline permission will be given subject to agreement on details; ii) Renewal Areas suitable for some new development or redevelopment; and iii) Protected Areas in which there is a presumption no development will be permitted. Planning would still be democratic and local. LAs will develop their own plans which may include enforceable design guides, and these plans will then be voted on and adopted or amended. But once adopted, they then will determine decisions so the current lottery of permission will disappear.

The problem of having so much uncertainty in our current decision-making – apart from the fact it means there is in effect no systematic planning – is that the uncertainty creates a substantial cost to developers. They need to build in a risk premium for any proposal to be viable, so far fewer are. Take away the uncertainty as to the decision and many more proposals will be viable: more houses will be built.

The second excellent proposal in the white paper is to abolish Section 106 Agreements and the Community Infrastructure Levy (CIL) and replace them as I suggested in this Planning piece (paywall) with a simple Infrastructure Levy (IL) fixed as a proportion of the sales price of the developments. The revenue from such an IL – as explained in the report Homes on the Right Tracks would be very substantial and more than sufficient to pay not just for proper new infrastructure to support new development but produce a much more generous flow of revenues for much needed social housing. A further interesting twist to the IL proposal is that 25% of the revenues should be spent in the neighbourhood of the development. One can see the attraction of this since one of the underlying supports of NIMBYism is the reality that if local communities permit new development they are, in effect, fined. There is a statutory duty to provide new services for the additional residents, schools, health care etc, but no guaranteed flow of revenues to pay for it. Local infrastructure is liable to become more congested and service quality get worse. Having 25% of the IL revenues go to the local community would offset for this.

Section 106 Agreements are individual one-off agreements negotiated by LAs with developers setting conditions for granting permission. They most commonly oblige the developers to provide a proportion of the new homes as so-called “affordable housing”. However, as I explain below, if one was trying to come up with the least efficient method possible of making housing more affordable then S106 Agreements would be a strong contender. Planners and politicians like them because there are no visible costs to public funds and it gives a sense of fulfilment to planners. Big developers like them because S106 Agreements are a huge drag on smaller developers; so they reduce competition and speed the monopolisation of the development industry.

The substantial transactions cost – to both developers and LAs of negotiating them – is only the tip of the iceberg of their costs, however. The real resource cost is hidden and comes in the form of yet a third source of extra risk and financing costs in development. The financial returns from any development depend on the form and details of the S106 Agreement but those are not known in advance. Bigger developers are better placed to manage this risk because they have both access to capital and to local planning officers and politicians as well as their own experts paid to double guess the system. Like the randomised decision making embodied in our use of “development control/management” the use of S106 Agreements, therefore, adds yet more risk to the development process requiring yet higher risk premia, so fewer projects are viable and fewer houses are built. Add to this the extra squeeze it puts on small developers and the resulting reduction in competition further reduces total building. Of course building fewer houses over time increases their price so reduces housing affordability in aggregate.

Yet a further positive proposal in the white paper is to require LAs to decide how much land to allocate to either Growth or Renewal Areas explicitly to take account of local housing affordability (as originally suggested in The introduction of price signals into land use planning decision-making : a proposal) as a “binding requirement”. The details will matter but at least if the standardised format for local plans is up and running, it will be possible to see how much land each LA is allocating for housing so the “binding requirement” will be potentially enforceable. As things are at present since there is no central record, and in many cases no record, of how much land is being allocated for development - any rules supposed to be used to guide the process are effectively useless.

The white paper also contains traps and unresolved issues, some of which are subject to consultations. Perhaps the most dangerous trap is the proposal to enhance design quality by having an accelerated approval process to ‘fast-track’ beauty (Proposal 14). While better quality design and place-making is an admirable goal the danger of this particular proposal is that is difficult to see how it could work without reintroducing by the back door, a whole new process of local piecemeal and no doubt partial and contentious “development control” under a new guise. Much better to have good design guidelines and some rewards for high-quality development such as a partial remission ex-post of the liability for paying the IL.

Another danger is that the IL may degenerate into a form of CIL, controlled by the whim of LAs (indeed, as with local plans, only about half of LAs bother to charge CIL at all and of those that do, many manipulate rates to act as a second line of development control). For it to work to improve the planning system and land value capture for the community’s benefit it is imperative that:

  1. its revenues are spent only on items nationally identified – basically supporting local infrastructure and social housing;
  2. revenues are ring-fenced from Treasury depredations. No point in providing an incentive to encourage local communities to accept new development if the Treasury may take it away next year in its block grant determinations; and
  3. the same rate is applied in all local authorities. The reason for this last is, of course, to provide an incentive that varies according to local housing shortages. Where housing is more expensive a uniform rate of  say 20% on sales value – will generate more revenue. If there is concern that this would be unfair, damaging the “levelling-up agenda”, than, much better than having rates determined locally, would be to have a fixed proportion of revenues paid into some central “levelling-up fund” which would be used to provide differential support for the poorest communities or regions.

There are still important issues related to the present planning system and its crippling effects on housing supply that are unaddressed in the white paper but it seems more constructive to welcome the positives. If the proposals in the white paper are implemented in a clean and transparent way many more houses will get built, homes will become more affordable and many jobs will be created.