Hansel Teo, Florin Vadean and Julien Forder
Local authorities (LAs) in England are responsible for providing Adult Social Care (ASC) services, which includes residential and nursing care as well as community-based care, including direct payments. A person is eligible for LA-funded ASC if they are assessed to have a sufficient degree of need for care and support and meet the conditions of a financial means test.
The central government allocates funding for ASC to LAs using formulae to help account for differences in local funding needs. This means, for example, that in two LAs with the same number of people but with different numbers of people who draw on LA-funded care and support, the one with more people who draw on care and support would get more funding.
The allocation formulae, known as the ASC Relative Needs Formulae (ASC RNF), distribute fixed amounts of funding across LAs based on socio-economic indicators of local area care need and financial deprivation. These indicators include, for example, the proportion of people above age 65 living alone and the proportion claiming pension credit.
How well RNF allocations reflect current levels of need for LA-funded ASC across local areas depends on having up-to-date data on socio-economic indicators. The ASC RNF currently used to distribute funding was developed in 2005/06. Allocations based on this ASC RNF are thus unlikely to accurately reflect current differences in social care need across LAs.
This research produced an updated set of ASC RNF by:
Our statistical analysis aims to relate differences in levels of LA-funded care across local areas with differences in local area indicators of care needs and financial deprivation. The sizes of these correlations are then used to construct the ASC RNF.
To do so, we analyse data at small area level, specifically, Lower Layer Super Output Area (LSOA). These are areas of between 400 and 1,200 households and contain around 1,000 to 3,000 residents. Because up-to-date small-area level information on ASC is not available, we used data collected in 2013 as part of previous study on the ASC RNF. This data contains information on the number of people in each LSOA that accessed LA-funded ASC during 2012/13 and was collected from a representative sample of 60 LAs. The 2012/13 data was scaled to reflect levels of LA-funded ASC in 2022/23.
The statistical analysis accounts for potential differences in care planning, commissioning and other local characteristics across LAs. We also account for the effect of differences in supply of care home beds and social care workforce on local ASC support levels.
Accounting for the role of supply factors as well as observed and unobserved differences across LAs is important. Not controlling for them may lead to formulae which over or understate the importance of care needs and financial deprivation indicators. This would, in turn, lead to allocations which do not accurately reflect local area ASC funding needs.
We estimated four models: 1) community-based services for people aged 18 to 64; 2) care home services for people aged 18 to 64; 3) community-based services for people aged 65 and over; and 4) care home services for people aged 65 and over.
The relationship between LA funded ASC support and measures of need, financial deprivation, and ASC supply reflected in the derived formulae are consistent with theoretical expectations. For example, the formulae allocate relatively more funding to LAs with populations having higher care needs and which are financially less well off.
We assessed the robustness of our results by considering alternative indicators of care needs and local area wealth. Overall, results from these alternative specifications were highly similar to our main findings. These suggest that the main estimates are not sensitive to our particular choice of indicators.
Nonetheless, we recognise the limitations of using ASC support data from 2012/13. This limitation can be addressed if up-to-date individual level data of people receiving LA-funded ASC is made available to researchers. Such data is currently reported quarterly by LAs.
Generating reliable allocation formulae depends on access to good-quality data and requires that certain assumptions are made during the analysis. Our approach of using data on observed levels of LA-funded ASC, known as the utilisation-based approach, was chosen for this analysis based on the availability of data and well-established statistical modelling. The approach assumes that observed differences in levels of LA-funded ASC reflect the differences in levels of underlying care needs and eligibility for LA support. This assumption may not be valid if, for example, levels of LA-funded ASC are influenced by LA commissioning practices.
An alternative approach, which directly defines people’s need for LA-funded ASC based on disability and financial indicators, is the normative approach. However, this approach has its own limitations. For example, it requires additional assumptions regarding the level and forms of care needs eligible for LA-funded ASC and the amount of those needs that are not met through unpaid caring arrangements. Moreover, routinely collected data at individual or small-area level on care needs and unpaid care are not readily available.
