An update on the GLA 2024-based demographic projections

This is a brief update on the timing of the upcoming release, what it will include, and some of the changes being introduced with the new projections.

When will the projections be released?

The projections will be published alongside the Draft London Plan and the London Strategic Housing Land Availability Assessment (our release is tied to these because the projections incorporate housing assumptions based on the SHLAA). The publication date for the Plan has yet to be announced, but we are currently working on the assumption of a July release. [note: updated housing-led projections are already available to users in London local authorities via the GLA population projection service]

In the meantime, we have released an updated set of trend-based projections as research outputs. These incorporate a number of changes from the initial set published last August and should be much closer to the final outputs.

What will be included in the publication?

Trend-based population projections for local authorities
Housing-led population projections for electoral wards and MSOAs
Household projections for local authorities, with variants based on alternative scenarios of future population growth and household formation.

What is new in the projections?

This will be the first full update of the population projections since the publication of the 2022-based outputs in August 2024, and the first release of household projections since 2021.

It will also be the first set of projections to incorporate data from the new London SHLAA – the first major update to the housing assumptions used since 2017.

There are some other important changes coming with the new round of projections, including:

Adjustments to the underlying population estimates to account for issues with official data for children
Introduction of improved fertility rate estimates and projection methods
Expansion of the housing-led population model beyond London

Adjustments to population estimates of children

Earlier this year we identified problems with the numbers of children in official population estimates. We believe that there are two separate underlying issues:

Underestimation of young children in the 2021 census.
Inflated estimates of annual net international migration of children from 2021-22 onwards.

These issues are similar to those that affected estimates over the course of the previous decade, though that case the impacts were mostly limited to local authorities in London, whereas the current problems appear to affect estimates across England.

It took us a long time to understand the original issues and to develop a reliable method for adjusting the estimates to mitigate the impacts on population and school roll projections. That experience proved useful this time around, helping us to identify and address the problems much more quickly.

My colleague Donal will be writing more about the issues and his work to create the adjusted series in an upcoming post.

New fertility rate methods

Izabel previously wrote about her work to make detailed fertility rate estimates accessible to the analytical community and she will be sharing an update on recent progress soon.

This data has already allowed us to implement new methods for projecting fertility rates in the models, with the 2024-based projections featuring two improvements:

Projected fertility rates are produced using a new time-series approach that provides a more robust and flexible approach to creating scenarios than before.
Age-specific rates for small areas (e.g., wards and MSOAs) now much more closely reflect local patterns.

Expansion of the housing-led population model

Perhaps the biggest change to the models is that the housing-led population model has been extended to cover areas beyond the Greater London boundary.

There are a few advantages to expanding the coverage of the model, but the primary reason for making this change now is to better accommodate the potential impacts of recent changes to the planning system.

Changes to the planning system in England

At the point when the current Plan was developed, the accepted approach for determining future housing need in England was to take projected growth in the number of households and add an allowance for any existing backlog.

For the majority of the country, housing need was assessed at the level of individual local authorities. For London overall need was calculated for the region as a whole, and the GLA allocated housing targets to local authorities based on their capacity to delivery additional housing, as determined by the SHLAA.

In 2024, the government introduced changes to the National Planning Policy Framework, including a new standard method for calculating housing need in England. In the new method, housing need is entirely independent of demographic projections and based only on the existing number of homes in the area and local affordability measures.

Under the old system, housing targets for the region of London and individual local authorities elsewhere had a level of inherent consistency with trend-based projections. As such it was generally reasonable to use trend-based population projections for these areas without making any additional consideration for the impact of delivering homes in line with assessed need. The role of housing-led projections was primarily to determine the spatial distribution of the population within those areas.

The new standard method results in need figures that are, for most areas, significantly higher than projected increases in population and households given by trend-based projections. In the case of London, the new standard method gives a need figure of approximately 89 thousand homes per year, whereas official household projections give annualised growth of under 30 thousand.

The wide disparities between indicative housing need figures and current projections has complicated the use of trend projections for determining future growth in London and elsewhere.

The expanded model

The approach we are taking to address this issue, is to use trend projections to determine the overall population of a much larger area than the administrative region of London – something that better approximates London’s true housing market area – and then to use housing-led projections to distribute population within this area.

This approach hinges on the idea that migration can be categorised into two distinct types:

Long-distance or international moves tend to be linked to:

Pursuit of higher education
Employment opportunities
Relocation to be closer to family

Shorter distance moves, often labelled “residential mobility”, are primarily the result of changes to:

Housing needs, e.g. linked to life events such as family formation or separations
Tenures, e.g. purchasing a home and moving out of private rental accommodation
Priorities for amenities, e.g. moves to be closer to schools and green space after having children

We can use this distinction to define a set of Housing Market Areas (HMA), intended to represent areas where most moves within them can be considered as “residential mobility”, i.e. primarily driven by factors related to housing, and most moves between them can be considered as long-distance migration, where other factors dominate.