Scottish Welfare Fund review: final report - data analysis appendix

Summary

This report covers the data analysis as part of the review of the Scottish Welfare Fund (SWF), or 'the Fund', drawing on:

• Analysis of routine quantitative monitoring data, collected by local authorities and collated by the Scottish Government
• Contextual data analysis of official sources including data from the DWP, data from food banks, survey data and population data
• Data provided by the Scottish Prison Service and the Scottish Public Service Ombudsman.

This is an appendix to the main report of the review, which draws together the findings of the secondary data analysis with the findings of the evidence review, survey of local authorities and qualitative research with applicants, local authority staff and other key stakeholders. It should be read and interpreted alongside that report.

Increased need and demand for the Fund

Providing a definitive measure of underlying need for the SWF is challenging – there are various indicators of extreme financial hardship, but none of these neatly align with the eligibility criteria for the SWF. Analysis of food bank, destitution and Scottish Household Survey data indicates that far more households access food banks than access the SWF but the proportion accessing Crisis Grants is considerably higher than estimates of destitution and of being in 'deep financial trouble'. Meanwhile, the National Institute of Economic and Social Research (NIESR) have estimated that destitution could increase by 30% in the next financial year. While the evidence reviewed identifies the changes to the welfare system as the root cause of increased food poverty and destitution, the Scottish Welfare Fund is identified as providing a desperately needed and vital safety net.

The data analysis shows increased demand for the SWF. The number of applications for Crisis Grants increased by 134% from 2013/14 to 2021/22, while applications for Community Care Grants also increased by 51% over the same period.

Purpose under pressure

The increased volume of applications to the SWF, and in particular the increased volume of repeat applications puts the original purpose of the Fund under pressure. The Fund is intended to address one-off need. However, repeat applications for Crisis Grants have increased from 56% of applications in 2014/15 to 80% by 2021/22, while repeat awards have also increased, from 49% in 2014/15 to 68% in 2021/22.

Challenges of delivering a discretionary fund in a consistent manner

The guidance on delivery allows for "extensive discretion" over how the scheme is delivered across areas. Data from local authorities highlights that different areas do take varying approaches to the operation of the Fund.

Assessing whether differences in the data for different local authorities reflect appropriate local discretion or whether they may have implications for fairness of process and outcome for applicants is challenging. There are certainly examples that appear to fall into the former category – for example, analysis of monitoring data showed wide variation in the level of referrals recorded by different areas, but as those making fewer referrals includes some locations with higher levels of successful awards, lower referrals may be associated with a lower perceived need for alternative assistance. It is also unclear the extent to which these differences in reported referrals reflect differences in recording practices rather than actual variations in referral levels.

However, a key finding from the analysis of monitoring data is that local authority is an important predictor of whether or not applicants are granted either Crisis Grants or Community Care Grants even after other factors (such as their reasons for applying, their personal characteristics, mode of application, etc.) are taken into account.

Assessing whether these differences reflect appropriate local discretion or whether they may have implications for fairness of process and outcome for applicants is challenging. However, there was some evidence of differences between areas in assessment and/or recording practices which seem unlikely to reflect different priority levels or local needs, including variations in the level of applications rejected as 'incomplete'.

Future funding

Expenditure on the SWF in 2019/20 (pre-pandemic) was 108% of the allocated budget. There is some evidence that those areas that have historically tended to over-spend their budget (pre-Covid) are also those local authorities with a higher than expected rate of applications to the Fund (based on population size, number of people on Universal Credit, and number of low income families. However, this pattern was not universal – a number of low or average demand areas also over-spent their 2019/20 budgets. Over a longer period of time, consistent overspending has tended to be more common in more urban and mixed local authorities, while consistent underspending of allocated budget has been more common in rural local authorities.

The latest data for 2021/22 showed an overall overspend – 115% of the budget was spent. There were also some extremely high overspends – 198% of the budget was spent in West Lothian, 182% in South Lanarkshire, 169% in Perth and Kinross, 167% in Edinburgh and 162% in Dumfries and Galloway. Yet some local authorities had still underspent. The relationship between overspending and underspending and outcomes is nuanced, so it is very difficult to predict the impact of increasing funding, in local authorities where outcomes are poorer but where there is currently underspending, for instance. The impact may depend on whether poorer outcomes relate to understaffing, or other organisational issues. If extra funding allows staff to take more time collecting evidence of need, outcomes may improve and may encourage less rationing in decisions.

It is also important to note that, from the qualitative interviews (discussed in the main report from the review) many previously under-spending local authorities expect to overspend in the current year.

Understanding outcomes

Considering outcomes for applicants, Crisis Grant applications have always been more likely than Community Care Grant applications to result in an award. However, award success rates for both grant types fell in the years prior to the pandemic (and remained lower in 2021/22). These falling success rates are likely to reflect multiple factors, including budget pressures, and the changing profile of applicants, with applicants with specific vulnerabilities (who are more likely to be successful) accounting for a smaller proportion of applications in recent years.

Pre-pandemic, there were significant differences in SWF application success rates between local authorities, even after controlling for other factors that might impact success. In other words, all other things being equal (household type, reason for applying, disability etc.), the local authority people apply for a grant in had a significant impact on their likelihood of success.

As discussed above, the relationship between success rates and budget is complicated. While budget is clearly one important factor, there are likely to be other elements, beyond budgets, impacting on differences in success rates between areas.

Analysis of other factors that impact on chances of a successful grant generally suggest that level of vulnerability is key, and that grants do appear to be targeted towards those with more recorded vulnerabilities.

The average award level for Community Care Grants has reduced over time, while the average Crisis Grant has increased. Pre-pandemic, there were large variations between local authorities in the size of the average Community Care Grant award. These appear to reflect, at least in part, differences between areas in the reasons for applying for grants. However, budget also appears likely to be a factor, with historically under-spending areas making larger average awards.

Crisis Grant award levels varied less between local authorities and there was no strong relationship with budget overspends.

Only a minority of applicants to either grant were reported as having been referred to another service provider for help – around a third in each case. There is variation in recorded levels and stated approaches to referrals and signposting between local authorities.