Planning - the value, incidence and impact of developer contributions: research
Independent research on section 75 planning obligations and other developer contributions mechanisms. The report brings together quantitative and qualitative evidence to inform our wider review.
3. Evidence from the Survey
3.1 The Questionnaire
We sent an online questionnaire to all 34 planning authorities (32 local authorities and two national park authorities) in order to understand if and how local authorities implement developer contributions policies; their attitudes to the system; and their assessment of its effectiveness—and particularly how much is actually raised. A draft of the survey was discussed at a roundtable of practitioners who worked in planning and housing and had particular knowledge of the developer contribution system and of the data that were collected. The questionnaire text is included in the report as Annex 3 (Sub-Annex 3A).
The survey aimed to collect detailed evidence from each authority on both qualitative questions (how they used the system and how effectively it works for them) and quantitative ones (the numbers of planning permissions; the proportion involving developer contributions; the amounts agreed for affordable housing and infrastructure and collected over the last three years).
The qualitative questions covered:
(i) Planning authority policies, plans and guidance, including whether they had formal policies on developer contributions; when their development plan was adopted; supplementary guidance; whether they used standard charges and what areas were covered;
(ii) Local infrastructure plans--how detailed they are; types of infrastructure included in developer contributions; and what proportion of infrastructure needs are covered by expected developer contributions;
(iii) Affordable housing--whether it is sought; in what forms; and how much is received;
(iv) Operational effectiveness--how delivery is monitored; whether what is agreed is delivered; and
(v) Overall effectiveness--challenges; difficulties in achieving agreement; and consequent planning delays.
In terms of numbers we asked for:
(vi) The number of planning applications and number of agreements;
(vii) The use of the different approaches;
(viii) The number of homes consented and delivered;
(ix) The number of affordable homes permitted and completed, and the proportions that came via developer contributions;
(x) The tenures of affordable homes permitted and completed;
(xi) The number in each tenure that came via developer contributions
(xii) The number of agreements for obligations other than affordable housing; the value of those obligations; and whether they were in the form of land, in kind or financial contributions.
Respondents were not asked to provide values for affordable housing contributions. These were estimated separately by the research team (see Chapter 6 for details).
3.2 Response rates
The overall response rate was 100%, which is extremely unusual in research of this type and for which we are very grateful.
However, many authorities found it challenging to complete all the questions in the surveys, both because of Covid 19 and because of lack of appropriate data. The pandemic meant that respondents often had far higher workloads than normal, and many were working under difficult conditions, usually from home. Many respondents said the circumstances made it difficult to liaise with colleagues in other departments, which was often necessary in order to answer questions in detail.
As importantly, respondents often did not have full access to the types of data we were requesting. Additionally, the data around developer contributions are held in different ways across authorities and it was often difficult to provide material in the form requested. Some authorities did not have collection systems which included all the data we were requesting. As a result, few authorities were able to answer all of our questions. Many clarified the problems they had faced in completing the questionnaire and added additional qualitative explanations where appropriate. Respondents' comments on the problems they faced are included in Annex 3, as is the question-by-question analysis.
3.3.1 Policies, plans, and guidance
Only three authorities said they did not have formal policies on developer contributions (with one non-response) (Table 1).
Table 1: Does your planning authority have formal policies on developer contributions?
Of the 26 PAs who stated they had policies on developer contributions in their development plans, 58% had put the current plan in place in 2017 or later. Importantly 70% had supplementary guidance on developer contributions and 40% had non-statutory guidance (including many who had both), mainly adopted over the last five years.
About two-thirds of authorities used standard charges for at least some developer contributions. Requirements did not usually apply to the whole PA area but to zones or more likely projects or sites.
3.3.2 The importance of developer contributions
Over the three-year period from 2017 – 2020, our survey reported 7.7 % of planning permissions included planning agreements. Just over two-thirds came under Section 75 of the Town and Country Planning (Scotland) Act 1997 and the rest under Section 69 of the Local Government (Scotland) Act 1973 (used mainly for smaller /simpler contributions). Although we asked about Section 48 of the Roads (Scotland) Act 1984, no PA reported using it during this period. The proportions of permissions accompanied by agreements seem quite low but are a considerable increase on earlier estimates which suggested less than one percent had included agreements.
Specifying requirements (notably for transport infrastructure specific to the site) using planning conditions -- which do not involve direct financial payments -- was mentioned by a number of authorities as an alternative to developer contributions in some circumstances.
Developer contributions generally fall into two main categories: affordable housing (dealt with above) and infrastructure. The latter encompasses not only 'hard' infrastructure such as roads, bridges and public transport, but also facilities for education, health and leisure, and other public and environmental amenities.
With respect to infrastructure, almost 90% of responding PAs had Infrastructure or Local Development Plan action programmes, and half had Infrastructure or Capital plans. Over two-thirds said their plans included costings. 42% of responding authorities said their infrastructure plans included both funding and timing, but a similar proportion said the documents covered neither.
The range of infrastructure included in the various programmes and plans was very wide (Figure 2). Almost all included education and nearly 90% covered roads, followed by open spaces and recreational facilities. On the other hand, medical facilities were hardly mentioned. Almost 40% of respondents thought their authorities had infrastructure needs which were not reflected in their plans or programmes, partly because needs were constantly evolving.
(all PAs with infrastructure programmes/plans; n=24. Multiple answers permitted)
The evidence on whether contributions to specific types of infrastructure had been agreed in each of the previous three years suggested somewhat lower figures. Seventy-five percent of authorities said they had agreed contributions for education, and the same percentage for roads. Just over 50% had made agreements for open space and somewhat under 50% for recreational facilities (Table 2).
|Schools and other educational facilities||16||15||14|
|Roads and other transport facilities||17||13||14|
|Sporting and recreational facilities||12||11||12|
|Public realm improvements||6||7||4|
|Medical facilities/ emergency services||5||5||4|
(PAs that had entered into agreements with developers in the preceding 3 years; n=20. Multiple answers permitted.)
Respondents were asked what proportion of infrastructure costs they expected to be covered by developer contributions in one year's time and in the next decade (Figure 3). Fifty percent of PAs responded to the question. Those that did said relatively small proportions of the costs in the coming year would be met in this way. They were far more sanguine about the next decade, with 11 of the 17 respondents saying 40% of infrastructure costs or more would be covered. It is fair to say that most respondents thought their estimates were only somewhat accurate; just one thought their estimate was very accurate. A number commented on how difficult such figures were to estimate.
(Question seen by all PAs but not all responded; n=17)
3.3.4 Affordable housing
Affordable housing was undoubtedly the most commonplace form of requirement and of contribution. Eighty percent of respondents said they had policies to deliver affordable housing. Reasons for not having such policies among the 20% who did not have them included lack of identified need, viability, and the fact that the council itself met the requirement.
Those with policies mostly allowed all three means of delivery from developers: provision of land, completed homes and commuted sums, depending on circumstances at the individual site. Almost two-thirds of those with policies expressed their requirements as a proportion of new homes, using either the national 25% policy or their own Housing Need and Demand Assessment proportions. However 20% set no requirement. Requirements in all but one case related to residential developments rather than commercial. PAs set development-size thresholds below which no affordable housing contribution was required, which ranged from 2 to 50 units. Mostly it was social and/or midmarket rental that was to be delivered.
In the main the contributions required took no account of the extent of national subsidy. However, a small number of authorities mentioned that subsidy meant they did not have to rely on developer contributions alone to provide the amount of affordable housing their authority needed.
In terms of numbers: of the roughly 110,000 homes permitted over the three-year survey period, almost 30% were affordable units. Some 8% - 10% of all homes permitted were affordable homes delivered through developer contributions. Table 3 shows that in the areas with the highest house prices (upper quartile), about 25% were affordable units via developer contributions, while in other areas average proportions were generally 5% or less. This reflects both the concentration of development overall in a few high-priced PAs, and the relative ease of negotiation in those areas because developers are keen to get on site.
|Price quartile 1 (data for 5 of 8 authorities)||4%||0%||0%|
|Price quartile 2 (data for 3 of 8 authorities)||4%||2%||32%|
|Price quartile 3 (data for 2 of 7 authorities)||5%||5%||2%|
|Price quartile 4 (data for 6 of 9 authorities)||18%||22%||24%|
Source: Survey, Q39 (number of housing units permitted per year) and Q45 (number of affordable units permitted to be delivered by DCs)
With respect to tenure, around 70% of all affordable homes permitted were to be social rented properties with a further 22% being mid-market rent. There were hardly any shared ownership dwellings consented, and just under 10% were either partial equity or discounted market sale. Among all affordable homes completed, there were rather more social rented units and rather fewer for mid-market rent.
Looking just at the affordable housing provided through developer contributions, the tenure mix was very different. Contributions were involved in little more than 20% of both social and mid-market rented properties, but 93% of discounted market sales and all of the tiny number of shared ownership units.
About 40% of PAs answered our question on how long it takes to deliver affordable housing. The answer--an average of over two years and a maximum of four--does not suggest that such homes are delivered early to help developer cash flow, as is often suggested.
We examined whether permissions for new homes and affordable homes (both per thousand population) reflected house price levels and/or changes in house prices, to get some idea of what might determine activity rates. On average, 4.4 new homes of all types were permitted per thousand population in areas with house prices in the lowest quartile over the last three years, as compared to 7.1 in areas with prices in the highest quartile. This suggests some positive relationship between house prices and permissions, although there are variations across the three years. The relationship with increases in house prices, while similar, is less clear cut.
Turning to affordable homes delivered via developer contributions, output is much more strongly concentrated in higher-priced areas except for a clear anomaly in quartile 2 in 2019/20 (Table 3.3).
3.3.5 Operational efficiency
Over 70% of authorities said they had a dedicated team dealing with negotiating agreements, and 50% had a team that dealt with monitoring. Over 50% used site visits to monitor delivery but there were a wide range of additional/alternative approaches.
Around a quarter of PAs said that contributions were always delivered while over 60% they were mostly delivered. One PA had had none delivered. Authorities suggested that maybe up to 15% of agreements involved requests for variation – mainly because the scheme itself had changed. Most such requests were granted.
3.3.6 Specific Challenges
Finally, respondents were asked about the biggest challenges they faced, and this was a question they clearly wanted to answer. The two issues that dominated were viability and inadequate potential for developer contributions to achieve the required funding: 51 out of the 90 responses referred to these issues (Table 3.4, shaded rows). Negotiation difficulties were also important, mentioned by 40% of respondents.
|Viability issues for developers||21||78%|
|Getting enough contributions to deal with the impact of cumulative developments on infrastructure needed||18||67%|
|Delays to site starts and therefore payment of contributions||14||52%|
|Land/development market not strong enough to support what is needed||12||44%|
|Constraints arising from the five tests||8||30%|
|Total PAs responding||27|
(All PAs that had entered into agreements in last 3 years or previously; n=27. Multiple answers permitted)
A number of respondents also raised concerns about particular aspects of the current system. Three issues stood out:
i) Concerns about infrastructure costs, which are rising faster than the indexation metrics and are extremely difficult to predict. Education costs in particular seemed to be undervalued: authorities said actual costs were higher and rising more rapidly than the sums secured in contributions;
ii) The need to better align planning for new homes with transport and education estate strategies. Development should use existing capacity more effectively, and local authorities need to forecast future revenue funding requirements arising from infrastructure provision;
iii) The timing of consideration of developer contributions: rather than being decided when permission in principle was granted, in the Scottish system they were considered far too late in the process.
The 100% response rate to our survey was both unexpected and highly appreciated. However, respondents faced major challenges providing the range of data requested. Some of these problems were due to the pandemic, which meant respondents were not always able to give us all the details we requested. More fundamentally, though, it was clear that many authorities did not collect the specific data we required. In other cases, the information was only contained in individual planning files and proved to be too difficult to extract.
Despite these shortcomings, the evidence they were able to provide presented a very clear picture of how the system was operating and what was achieved across the country. Respondents also often provided additional information to help cover gaps.
The survey suggests that contributions are most regularly used for affordable housing and then for transport and education infrastructure but with instances of a wide range of contributions for other purposes, notably open space and sports facilities.
Developer contributions for affordable housing are well understood and relatively easy to negotiate – in part because there is a national standard. Some PAs said that they do not require developers to deliver affordable housing, sometimes because the evidence does not support an affordable housing requirement so they do not need such contributions, but more usually because they feel that requiring them would either make schemes non-viable or would drive away development.
Developer contributions for infrastructure are far less embedded in the system. Most PAs do have infrastructure action programmes and 50% have plans in place, but authorities generally recognise that because of viability constraints, contributions will only make a small dent in infrastructure requirements in the following year. Even so, authorities expect developer contributions to cover a higher proportion of requirements over the next decade.
The numbers presented make it clear that developer contributions are only negotiated for a small proportion of all planning permissions. Even so, survey evidence showed that around 30% of the affordable housing permitted is provided with the support of developer contributions (as well as government subsidy). Moreover, the vast majority is provided in higher valued areas.
Overall, the survey responses suggest that the developer contribution system is well embedded and is growing in importance. It shows that using contributions for affordable housing is well understood and generally accepted. The evidence with respect infrastructure suggests that PAs are comfortable with contributions directly related to the site and demand arising from that site.
This section addresses the information which we requested from the Planning Authorities. It does not address the issues around the valuation of these contributions. This is addressed in section 6.
There is a problem
Thanks for your feedback