There is no single factor governing the rate of delivery of a site. Considerations include site size, market strength and pricing
The annual number of residential consents granted has exceeded 300,000 in England for the past two years. This level needs to be sustained and even pushed higher if the target rate of completions is to be achieved over the long term. But, recently, there has been a downturn in the volume of permissions; is this merely a blip or a trend to be concerned about?
With the annual rate of consents exceeding 300,000, there is growing political focus on delivery rates, and more attempts to ensure consents are built out, notably through the Letwin Review. Our previous research papers have highlighted the need for a greater proportion of planning permissions to be granted in the least affordable areas, where demand is greatest.
In this year’s paper, we go beyond that analysis and look in more detail at the factors that impact rates of delivery. We also consider whether the current trend towards planning permission being granted on ever-larger sites will improve build-out rates.
David Jackson, Head of Planning
Sizing up the pace of delivery
The pace of housing delivery has become a key topic, particularly in light of the Letwin Review, which looked at the gap between the amount of land receiving consent and housing completions. As a consequence, it has become an issue of increasing scrutiny both in the preparation of local plans and at the planning application stage.
Despite this, there is no established methodology for assessing how quickly a site can deliver new homes. Market capacity (in an active market) is often approximated using the rule of thumb of one sale per outlet per week. The overall rate of delivery is then estimated by multiplying the number of outlets that can be supported by the site and adding the affordable housing scheduled to be delivered.
However, our analysis shows that the reality is more complex. The strength of the local housing market, the number of sales outlets operating on site, and the pricing relative to the local market, all influence the absorption of open market new homes.
The Letwin Review identified that diversity in product type and tenure mix can also boost delivery rates. That said, our previous analysis showed that it is not the only answer (Spotlight 2018: What next for housebuilding?).
There are trade offs between pricing that enables high delivery levels and the land value uplift that can be captured through developer contributionsSavills Research
Variation in delivery
Our latest analysis covers 41 large greenfield sites with capacity for at least 500 homes. These are located across England, Scotland and Wales, in markets of varying strength. We have tracked annual completions, covering open market sales, purpose-built rental stock and affordable housing tenures.
The average delivery rate from these sites is 145 homes per year, but there is a wide variation around this average. So, while 30% of the sites deliver more than 175 homes per year, three have delivered at more than double the average rate. Six sites, however, have delivered at less than half the average rate.
Delivery rates Distribution of completions (all tenures)on sites with capacity of more than 500 homes
Source: Savills Research
Scale and sales rates
There is no correlation between site size and the rate of delivery; sites with greater capacity and, hence, more physical potential to have multiple outlets, do not automatically result in higher delivery rates.
Equally, there is little clear evidence that a rule of one sale per week per outlet can be applied across all markets. Comparing Berryfields and Priors Hall, two sites with similar total capacity, shows the variation (below). While the trendline at Berryfields follows the benchmark line closely, at Priors Hall the lines diverge.
Berryfields (Aylesbury) sales
Source: Land Registry
Priors Hall (Corby) sales
Source: Land Registry
At this development, the highest monthly sales rates were achieved when there were four outlets active on site. Adding more outlets had no impact on increasing the number of sales across the site. This shows that there is much more that needs to be taken into account when assessing delivery rates beyond the total capacity of a site and the number of outlets it can support.
For open market sales, the strength of the local housing market and the demand for homes will be a key factor (chart below). Our research shows that for sites where sales rates are at least 50 homes per quarter, this has been achieved in areas where housing is least affordable (with affordability providing a proxy for demand).
Sales rates Peak quarterly sales rates and housing scarcity
Source: Land Registry, ONS
A further factor to consider is the pricing of new homes relative to the local market. The sites in markets that are selling high numbers of new homes tend to be priced at a discount to the average price of homes in the local second-hand market. Our previous research has shown that three of the highest delivery sites in higher-demand areas in 2017 were, on average, priced at a discount of up to 15% from local market pricing per square foot (Spotlight 2017: On track to solving the housing crisis?).
There are, therefore, multiple considerations that need to be taken into account when assessing the potential delivery rate of any site and particularly larger sites. There is no single rule that can apply across all sites; instead, consideration needs to be given to multiple factors, such as the level of housing scarcity in the area, the strength of the market, and the proposed pricing of the scheme. The strength of the local market is likely to have the greatest impact on build-out rates.
The analysis of the likely rate of completions will also need to be carried out in conjunction with the viability testing at local plan stage. This analysis indicates a potential trade-off between pricing that allows for higher sales rates and the amount of land value uplift that can be captured through developer contributions.
Read the articles within this publication below