In April 2020, the Scottish Government takes over responsibility for £3.4 billion of social security payments. Most of this money will be handed straight back as the Scottish Government is paying the Department of Work and Pensions (DWP) to continue to manage devolved benefits while it sets up their Scottish equivalents. The chart below shows spending in 2020-21 by the agency administering payments. (Local authority figure includes council tax reduction).
This blog looks at the scale of Scottish social security, arrangements for funding it and the challenge of forecasting spend for this demand led budget.
Social Security Scotland, set up in September 2018, will administer less than 4% of devolved benefit spend in 2020-21. This will change over the next few years as Scottish disability benefits start. It’s unclear quite how quickly it will change because, while we know when benefits will start for new claims, the Scottish Government hasn’t published a timetable for transferring the existing 500,000 cases from the DWP to Social Security Scotland. The intention is to complete in 2025, with the majority being completed by 2024.
Development and running costs
The cost of developing the Scottish system is twice what was originally envisaged in 2017 and will take seven rather than four years.
The revised business case shows that ongoing running costs of Social Security Scotland are also higher than previously estimated. These are likely to be in the region of £253 million in 2024-25. (Operational and DWP agency agreement costs in the chart below). This is a third higher than the £190 million upper estimate given in 2017 (adjusted for inflation). The Scottish child payment (which will pay out in the region of £180 million of benefit each year) will add £18 million per year to running costs – a higher ratio of operating costs to benefit spend than the estimated 5.3% for all Social Security Scotland benefits.
Funding
Most of the cost of benefits will be covered by a funding transfer from the UK government in the form of a block grant adjustment (BGA).
The idea is that the BGA reflects what would have been spent on benefits had they not been devolved. As illustrated below, the first year’s calculation is based on the spend in Scotland adjusted in line with the change in spending per capita in England and Wales. After that, the calculation is based on the previous year’s BGA again adjusted for changing spend per capita in England and Wales.
This means that as years go on the baseline, based on spending in Scotland, becomes a smaller proportion of the BGA. This means that the Scottish Government pays for its own policy decisions but also that the Scottish Government will fund the difference if take-up rates grow more quickly in Scotland than in England and Wales. This is because the BGA is calculated on how spending per head changes in England and Wales not Scotland.
While this might raise some concerns the chart below shows that if you do the calculation using existing data on benefit spending in England and Wales and Scotland going back to 1996, it shows little difference between the BGA calculation and actual spending in Scotland. This suggests there isn’t a systematic level of underfunding created by the calculation itself. However, given the large scale of the total transfer, even a relatively small divergence can be higher than total spending on some of the smaller benefits.
Block Grant Adjustment Reconciliation
To complicate matters, the BGA is based on Office for Budget Responsibility (OBR) forecasts and later reconciled to outturn spending. This reconciliation process could lead to the Scottish Government returning money to the UK Government or gaining additional grant. The normal timeline is illustrated below using the 2021-22 budget:
It doesn’t help that in this first year of large scale social security spending the delayed UK budget has meant the start of the process is slightly different (which is why the chart uses 2020-21). Provisional BGAs were provided in February and these will be updated in March after the UK Budget.
This BGA process is not a reconciliation to spending in Scotland. It’s a reconciliation from forecast to actual change in spending in England and Wales. The budget for actual spending in Scotland is forecast by the SFC.
Spending and SFC forecasts
Forecasting the budget accurately is particularly important for demand led budgets but forecast error is unavoidable.
So far, over-estimating in one benefit (carer’s allowance) has made up for underestimating take-up of the best start grant, and the numbers involved have been small (see Table 4 in SPICe briefing for Social Security Committee meeting 27 February). However, the impact of forecast error on the budget will increase as the scale of the benefit spend increases. There’s no reason to think that forecast error differs by the size of the benefit and disability benefits have been notoriously difficult to forecast in the past.
Additional Spending
While much devolved benefit spend continues along UK policy lines for the moment, the Scottish Government has made some changes. In 2020-21 these are likely to add at least £80 to £90m to social security spending above what is transferred from the UK government.
Estimated additional Scottish Government funding of Social Security Scotland benefits in 2020/21
Benefit |
Additional funding |
Notes |
£6m |
SFC forecast £6m additional spend from policy differences from child DLA. Assumes introduction in September 2020 |
|
£21m |
No DWP equivalent. Assumes introduction in December 2020 |
|
£39m |
No DWP equivalent |
|
£16m |
Deducting estimated £2m DWP spend from £18m BSG forecast spend in 2020/21. |
|
£1m |
No DWP equivalent. |
|
£2m |
No DWP equivalent |
|
Total |
£86m |
SPICe calculations based on SFC forecasts. There will also be additional spend on local authority payments. BSG is funded under Barnett not BGA.
Unknowns and uncertainties
The inevitable uncertainty over both the BGA and SFC forecasts will make it difficult to know how much is available each year for additional policy commitments.
The Cabinet Secretary for Social Security and Older People has referred to an average 3.7% forecast error for UK disability benefit spend – around £130 million. This is similar in scale to the entire forecast spend on Social Security Scotland benefits in 2020-21. Whether disability benefits are over or under-estimated could have a huge impact on the policy choices around the smaller benefits.
We don’t yet know the detailed policy costing for adult disability benefits and this is likely to add a significant amount to future budgets. Even a small change can have a large budgetary impact. For example a 1% additional cost to Personal Independence Payment (PIP) in Scotland would be around £16m – similar in scale to the total cost of the best start grant.
Another unknown is how much the launch of ‘Scottish PIP’ will boost take-up. The new system is intended to feel very different to the DWP system and there’s evidence of underclaiming of current DWP disability benefits. The SFC suggest there will be a spike in new claims for the child disability payment, and it is not unreasonable to expect something similar for the adult disability benefits.
The SFC haven’t forecasted this yet as they haven’t had the policy detail from the Scottish Government to allow them to do so. SFC forecasts are likely to be significantly revised – and likely upwards – as they start to calculate in the impact of Scottish Government policy.
A different scale…
The vast difference in scale of the different benefits – from £1.5 billion for PIP to £1 million for the young carer grant makes it less likely that increased demand or forecast errors in one area would be evened out by decreased demand and forecast errors in another. This reduces certainty around the affordability of policy changes.
Camilla Kidner
Senior Researcher
Social Security
SPICe