Planning - the value, incidence and impact of developer contributions: research

6. The Value of Developer Contributions

6.1 Introduction

In this chapter we explain the methods and data used to estimate the value of developer contributions where these have been made under formal legal S75, S48, and S69 agreements. Using evidence from the local authority survey (see Chapter 3) on the agreements made with developers including on the numbers of new affordable homes agreed as well as agreements related to infrastructure provided in kind or in cash by developers, we first estimate the value of the contributions made for affordable housing in the three years from 2017-18 to 2019-20. We then use the more limited information that we were able to obtain through the survey on the contributions made by developers towards infrastructure to estimate the value of the latter contributions agreed in 2019-20. . Together these two calculations enable us to make an estimate of the total amount of contributions agreed in the year 2019-20 covering both affordable homes and infrastructure...It is not possible to estimate the value of contributions that arise from developers meeting specified planning conditions as information on what is implied in terms of financial commitment to comply with the conditions will not be recorded.

6.2 Methods for valuing contributions to affordable housing

The following are the steps taken to make the estimate. The method follows the principles adopted in valuing contributions in England (Crook et al, 2016; Lord et al, 2020) and also draws on helpful advice from Rettie & Co, developers, local authorities and housing associations as well as government officials

Contributions are of three kinds:

(i) Developers provide serviced land to affordable housing providers (principally registered social landlords) with the contribution being the value of the serviced site less any payment from the association. Our case studies, roundtables and interviews with key informants suggest that the land is usually transferred at nil consideration (but sometimes for a nominal £10k a plot).

There is no published data on land prices in Scotland so we had to calculate land prices ourselves. We did so by valuing the nil consideration as the difference between the residual value of land (i.e., what a developer would pay to a landowner) when all homes are market ones and when only 75 percent are market homes. The difference between these two residual values represents the 'contribution' made by the landowner when a 25 percent affordable housing requirement is in place. This make no allowance for any transfer of value to developers or landowners or indeed infrastructure arising from the availability of subsidy, except to the extent that we have calculated what the value might be if £10k is paid for each plot.

We calculated the residual value for the full market site as the gross development value derived from a 105 dwelling 3-hectare site based on average house prices for newly built homes in each PA less all the costs of building the homes, servicing the site, all fees, marketing costs and financing and developers' profit. We calculated the residual value of the site with a 25 percent affordable housing requirement as the gross development value from 79 new homes (i.e., 75 percent of the site) less all the costs of constructing and selling these plus the costs of servicing the site for the 26 affordable homes.

(ii) Developers sell completed units to registered providers and local authorities that can be used for social rent, mid-market rent, shared ownership, shared equity or discounted market homes. In these cases, the developer's contribution is the difference between the market price for the completed homes and the price paid by a registered provider. We calculated what registered providers could in principle pay for rented units by estimating the discounted net rent over a 30-year period as the debt the provider could afford to repay and adding to this the grant for a 5-person dwelling, taking account of the variations between each local authority in rents and grant. We recognise that many registered providers could pay more than our estimate to buy by borrowing more cheaply or using their reserves but we have no information on either of the latter. The numbers of shared ownership and shared equity were so small we simply included them as equivalent to mid-market rental units.

(iii) Developers have sometimes been required to include market homes to be sold at a discount from full market price as part of their affordable housing obligations. We assumed in our calculations that discounted market sales were made at 70 percent of market prices for newly built homes in each local authority.

(iv) The above calculations provided us with an estimate of the contributions per dwelling agreed. These were then used to calculate the total amounts per PA for each type of contribution using the data provided in our survey on the numbers agreed and whether these contributions were in the form of serviced land or completed units (and the tenure of the latter). Because not all PAs were able to provide these data we grossed up the totals to provide an overall estimate for Scotland taking account of the 2021 mid-year population of responding and non- responding populations

(v) Commuted payments made by developers to local authorities as alternatives (or in addition to) the above. We asked all PAs to provide us with information on the sums agreed as commuted payments for affordable housing. Four were able to provide information on commuted payments towards affordable housing which had been agreed in 2019-20 totalling £1.8m. Five were also able to tell us about commuted payments for all types of affordable homes received in 2019-20: they received £8.5m in all.

6.3 Estimated value of developer contributions for new affordable homes in Scotland in 2019-20

Table 6 below shows the results of our estimates, based on the approach set out above and assuming land was transferred at nil consideration.

Table 6: Value of developer contributions agreed for new affordable homes in Scotland 2019-20

Type of contribution and dwellings (grossed up survey totals)	Estimated grossed up national total (£m)	Proportion of Scotland from top five local authorities#
Transfer of discounted land to registered provider for 2,700 dwellings*	82	45
Sale of completed units to registered providers for 1,150 social rented homes	161	44
Sale of completed units to registered providers for 505 mid-market rented homes	42	33
Sale of 180 market homes at discounted prices	15	44
Total	300	43
Commuted sum agreed with four LAs in 2019-20	1.8~
Commuted sum for all uses paid to five LAs in 2019-20	8.5~

*Note: all dwelling numbers rounded to nearest ten as these are grossed up numbers.
#Note: the top five local authorities in a ranking of authorities by total value of estimated contributions.
~Note: not grossed up.

If instead of developers transferring land to registered providers at nil consideration, the latter paid an average of £10k per plot the value of the developer contribution in the first row would be reduced to £54m.

Tables 7 and 8 (below) show our estimates for the years 2018-19 and 2017-18. We have no information from our local authority survey on the amounts of commuted payments agreed and received for these latter two years but taking the totals for discounted land and for completed units only we can see that the total sum agreed rose from £220m in 2017-18 to £300m in 2019-20 an increase of about a third over that period, reflecting a growth in numbers agreed as well as in the underlying financial variables used in our estimates including house prices and rents.

Table 7. Value of developer contributions agreed for new affordable homes in Scotland in 2018-19

Type of contribution and dwellings (grossed up survey totals)	Estimated grossed up national total (£m)	Proportion of Scotland from top five local authorities
Transfer of discounted land to registered provider for 1,900 dwellings*	66	48
Sale of completed units to registered providers for 920 social rented homes	147	48
Sale of completed units to registered providers for 220 mid-market rented homes	27	47
Sale of 130 market homes at discounted prices	11	47
Total	251	48

*Note: all dwelling numbers rounded to nearest ten as these are grossed up numbers

Table 8. Value of developer contributions agreed for new affordable homes in Scotland in 2017-18

Type of contribution and dwellings (grossed up survey totals)	Estimated grossed up national total (£m)	Proportion of Scotland from top five local authorities
Transfer of discounted land to registered provider for 1570dwellings*	57	51
Sale of completed units to registered providers for 790 social rented homes	131	50
Sale of completed units to registered providers for 240 mid-market rented homes	30	49
Sale of 30 market homes at discounted prices	2	50
Total	220	50

*Note: all dwelling numbers rounded to nearest ten as these are grossed up numbers

The five authorities contributing approximately between four and five in ten of all contributions are in the central belt: City of Edinburgh, East Dunbartonshire, East Lothian, East Renfrewshire, and Midlothian, where high land values can support the provision of affordable homes (and still leave room for infrastructure contributions where they have been sought by the PAs) while ensuring the developments remain viable. This is consistent with the evidence presented in chapter 3 which showed very high proportions of contributions concentrated in the higher priced areas of the country. In these authorities the value of the developer contributions of serviced land amount to an average of 32 percent of the land value with planning consent in those authorities but without any developer contributions. Were all other PAs to have met the 25% requirement it would have taken 85 percent of land value. This reflects how little has been, and possibly can be, captured from land values in the rest of Scotland were the objective to support a 25 percent affordable housing contribution. Indeed, our estimates produce negative land values in several of these authorities indicating that if affordable housing is required the percentage must be much lower than a quarter of residential sites. Taking the figure for 2019-20 on the total value agreed of £300m (apart from commuted payments) this equates to an average of £9.3m per PA.

6.4 Valuing contributions to infrastructure

We have been unable to estimate the level of all contributions to infrastructure directly as PAs (for reasons explained in Chapter 3) did not retain this information in the form requested or could not access the data we needed. Our intended approach was to collect information under types of contribution (land, in kind contributions, and direct financial payments) and for each type of contribution (i.e., for education, open space etc.) together with totals but this proved impossible.

We are therefore using the limited evidence that we received from about a third of all authorities to make some estimates under heroic assumptions.

There are three types of developer contributions towards infrastructure:

The first is the value of the land transferred to PAs (e.g., on which to build a school). To estimate this, we needed information of the amounts of land transferred and land values. However, we were unable to collect information of the hectares of land transferred from PAs from our survey. Therefore, we could not use the residual land values we had calculated for housing to make an estimate.

The second type is the value of financial contributions for agreements which provides the value directly.

Third is the value of in-kind agreements e.g., the transfer of new school classrooms built by the developer. This can be estimated indirectly by assuming that the average value of financial contributions per agreement for each type of contribution is equivalent to the value of the in-kind contribution (this is the approach taken in similar work in England based on detailed analysis of samples of agreements in PAs there).

Unfortunately, PAs were generally unable to break down the numbers of agreements by contribution type, did not have records available on the in-kind contributions agreed, nor on the land given by developers. Hence, we have been unable to place a value on in kind infrastructure contributions.

We also have no evidence on the amounts involved in meeting planning conditions where developers are required to ensure infrastructure is provided on the site where permission has been granted which, while technically not contributions, also impact on what is possible to capture from land values where the developers pay for, or provide in kind, the infrastructure specified in planning conditions.

6.5 Our estimate of financial contributions to infrastructure

Some PAs were able to provide us with data on the value of financial contributions agreed for the most recent financial year.

As is shown in Table 9 the total value agreed by the 13 PAs who had data with both the numbers of agreements and the value showed that there had been 483 agreements involving financial contributions to infrastructure which were valued at £80.8m or £6.2m per PA. (with a range of £141k to £31.2m). Fourteen PAs told us about the total number of agreements (563) but only 13 could provide evidence on their financial attributes. It is notable that the same five authorities that agreed a large proportion of affordable housing contributions (see above) also secured 79 percent of the reported value of infrastructure contributions out of the 13 PAs who were able to provide us with data on what was agreed.

Table 9: Value of financial contributions to infrastructure in 2019-20

Financial contributions to Infrastructure	Total sum agreed or paid to PAs providing information (£m)	Sum per PA providing information (£m))	Grossed up total for Scotland (£m)
Contributions agreed with 13 PAs in 2019-20	80.8	6.2	179
Contributions received by 9 PAs in 2019-20	54.5	6.1	186

Only 9 PAs were able to estimate the financial contributions received in 2019-20 (which cover funding from earlier agreements made in the past) for infrastructure. These nine PAs estimated they received £54.5m in total for infrastructure (an average of £6.1m per PA).

Using the same population-based method that we used for affordable housing we grossed up these totals to the whole of Scotland. Of course, because we only have data from a third or less of all PAs the estimate has a much lower confidence level. However particularly because it includes most of the largest, most active, authorities it is probably in the right ballpark.

6.6 Conclusions

Taking the grossed-up figures for affordable housing and infrastructure agreed together – the overall figure is £479m – i.e., nearly half a billion pounds. This equates to a contribution from developers worth £30,500 for each private sector house completed in Scotland in 2018-19. In England in 2018/9 the comparable figures were roughly £7bn of which £2 billion were for infrastructure and worth £50,400 per private house completed (but in England all contributions to infrastructure were valued). Per head of population this suggests Scotland is agreeing around 85% of that agreed in England. Moreover, the estimated split between affordable housing and infrastructure is around 65:35%, while in England it is roughly 70:30. These suggest that the systems are working in relatively similar ways.