Scottish Welfare Fund Statistics: Annual Update: 2021-22

About the Data

How the data is collected

One month after the end of each quarter, local authorities are asked to submit an XML file containing the information as set out in the data specification.

The data specification for the Scottish Welfare Fund and guidance to help local authority officers to record information are available at: the Scottish Government SWF data specification and guidance page.

The data sent to Scottish Government each quarter should include all applications where an initial decision, Tier 1 review or Tier 2 review have been completed, plus any updates to previous cases (e.g. were information within a case has been revised, or a case has reached a different stage). To take account of delays between applications being received and decisions made, and retrospective changes in award values, the data cuts that are sent to Scottish Government each quarter should be sent at least a month after the end of the quarter that will be reported on.

This data is then uploaded to the Scottish Government's ProcXed^[8] website. Once the data has been validated by our software, the data is passed over to the Scottish Government's Communities Analysis Division.

Scottish Government also collects informal, monthly management information through a separate process as described in our recent management information publications available at the Scottish Government monthly management information web page.

Revisions

Each updated publication of statistics will include revisions to figures for previous quarters, with more recent quarters being subject to a greater degree of revision than more distant ones. There are several reasons for this:

(i) the cuts of data received by Scottish Government each quarter will include retrospective changes to past applications. For example, where the actual amount spent on an item was different to the amount initially awarded, this expenditure will be updated in the new data cut.

(ii) the cuts of data received by Scottish Government only include information about applications that have at least reached the stage of having an initial decision made. Some applications may therefore be received by the local authority in one quarter and decided in the next quarter. Scottish Government would only receive details about these applications in the next quarter's data cut, at which point the application will be added into the previous quarter's application statistics retrospectively.

(iii) Tier 1 and Tier 2 reviews can lead to changes in final outcome and final award amount. Where a case has been reviewed, expenditure is assigned to the quarter of the initial decision rather than the review date.

Processing times

Initial Processing Time is the number of working days between the date all information was received and the initial decision. Our count of working days counts the number of Mondays to Fridays between these two dates. No allowance is made for local holidays and bank holidays. If the date all information was provided is missing, the application date is used instead.

Households

We collect information on all people within each application. To group people into household units we filter the people table to contain only the main applicant and their partner. We ensure that there is no more than two people for each application. We generate a unique household reference based on the hashed National Insurance Number, gender and date of birth of the main applicant, and if applicable, their partner.

Repeat applications and awards

An application is defined as a repeat application if the household made another Community Care Grant/Crisis Grant application within the 12 months of the current application. Similarly a repeat award occurs if an award was made within 12 months of the current award. While there is no limit on the number of Community Care Grants an individual can receive in any one year, Crisis Grants are normally limited to three in a rolling 12-month period.

Payment times

The time taken to receive a payment was calculated as the date all information was received to the date of the last payment. If the 'all information date' is missing, the date the application is received is used instead. Last payment date includes payments made as part of reviews.

Data Quality

To ensure the data we receive is of the highest quality, we have asked all IT providers to implement the data specification as fully as possible. The data specification includes a number of validations. These validations are provided through an XSD schema and have also been implemented into the ProcXed system. Once local authorities submit data to the ProcXed system, a number of validations are triggered and warning messages appear. Local authorities may re-submit data to fix these errors or they can comment them to explain why they believe the data to be valid. On submission of the data to the Scottish Government, each local authority is sent a detailed automatic report showing the contents of the data and any remaining errors.

To ensure sufficient data has been submitted, we cross check all submitted data with the monthly management information returns submitted to the Scottish Government (Chart 7, Chart 8). The automatic reports sent to local authorities contain charts which allow them to easily see if there is a discrepancy between these two data sources.

Known data issues to March 2022:

• City of Edinburgh experienced a software issue in March 2020 that has caused some crisis grant awards and expenditure from March 2020 to be recorded in April 2020 in the quarterly data extract supplied to Scottish Government. Comparison of the quarterly data to monthly management information (Table 74) indicates that around 1,335 crisis awards and £141,000 associated expenditure made in March 2020 has been shifted into April 2020 in the quarterly data extract used to produce this publication. This means that in Edinburgh in Jan-Mar 2020 around 33% crisis awards and 38% expenditure are missing, and in Apr-Jun 2020 around 17% of crisis awards and expenditure should have been recorded in the previous quarter. Scotland totals will also be affected. It is not currently possible to amend case details so that they appear against the correct month/quarter/financial year.
• In March 2020, some COVID-19 related Crisis Grant applications received by Glasgow City were recorded as CCG applications in the quarterly data extract supplied to Scottish Government. Comparison to the monthly management information (Table 73, Table 74) supplied separately by the local authority indicates that in March 2020, around 1,000 applications, 400 awards and £60,000 associated expenditure has been recorded as CCGs rather than Crisis Grants. However, this is difficult to quantify exactly due to ongoing and pre-existing discrepancies between the monthly management information and quarterly extract. These issues will also affect Scotland totals. From April 2020 onwards the issue of COVID related crisis grants being recorded as CCGs appears to have been resolved.
• Scotland level figures are affected by specific issues described for Edinburgh and Glasgow. There may be additional issues with data quality related to COVID-19. For example, comparison of management information (Table 73, Table 74) and figures derived from quarterly data extracts indicates discrepancies in numbers of applications, awards and expenditure for several local authorities. In many cases, the quarterly extract included fewer CCG and/or crisis grants than the monthly figures. This could be due to delays in applications being processed, or awards being kept as 'pending' or 'in principle' on local authority systems until they can be delivered/installed etc. However, overall at Scotland level, the monthly management information and quarterly data extracts have similar figures for CCG applications (-3% in the quarterly extract compared to the monthly management information) and expenditure (-1%) but there is a slightly larger discrepancy in CCG awards (-5%). At Scotland level, the monthly management information and quarterly data extracts have similar figures for Crisis Grant awards (-2% in the quarterly extract compared to the monthly management information) and expenditure (-1%) but there is a slightly larger discrepancy in crisis grant applications (-5%). Any issues caused by processing or delivery/installation delays should be resolved in future updates to the publication as more data extracts are received.
• Midlothian were unable to supply information on returns in February and March 2022. This has meant that their data for this period has been estimated, where possible, based on their January 2022 data, and on Scotland-level changes. This information will be updated in subsequent releases.

Missing information/delays in receiving information

An application is included in the quarterly data extract to the Scottish Government only when an application has been decided and when a payment has been made. For Crisis Grants, applications are less complex and payments are made quickly, resulting in almost no lag between cases being decided and being included in the data extract. However, for Community Care Grants, a large number of items may need to be ordered and payment may not occur until up to one month later, when invoices arrive. This results in a lag time between applications beginning in the local authority and applications being included in the extract to the Scottish Government. To rectify this problem, we are asking local authorities to submit data one to two months after the end of the quarter. This will ensure that, as far as possible, all relevant cases are included in the data extract.

Some local authorities may not have been able to enter all of their data in time to include in this publication. Late entry of data may mean that this data appears in subsequent quarters, rather than in the quarter when the case was actually dealt with. As dates are automatically system generated, late entry of data has a knock on effect on any calculation which involves processing times or payment times.

Some question responses may default to "Other" in some IT systems. This particularly applies for reasons for applications, rejections, reviews, review decisions and payment methods. We are currently working with IT providers and local authorities to ensure that the full range of questions and responses are available in all local authority IT systems. Some questions may not be completed at all. This includes the vulnerability and referrals questions for some Local Authorities.

Expenditure

Unless otherwise stated, all expenditure information in this publication is based on the date of decision. As such it should be regarded as committed spend, rather than actual spend. For example, a local authority may commit to purchasing floor coverings for a household and set aside money for this (on the decision date). In practice, it may be some time later before the local authority is invoiced and pays the money to the supplier (on the payment date). Overall, there is very little difference between analysing expenditure information using payment dates as opposed to decision dates. Where a case has been reviewed, expenditure is assigned to the quarter of the initial decision rather than the review date. This means that if the initial decision took place in 2016/17 but a review decision is made in 2017/18, all of the expenditure for the case is counted in 2016/17.

It is possible for the value of an award to increase or decrease between the decision and payment date. For example, local authorities may estimate the cost of carpets and the final invoice for these may be less than originally thought. This results in money being paid back into their systems. Where we receive updated information about this change in expenditure in quarterly data returns, the statistics for previous quarters are updated to reflect this. We also receive separate reconciliations information from some local authorities. Over time, the expenditure data in the quarterly monitoring may exceed monies actually spent if reconciliations are not applied. Most recently, reconciliations have been applied to South Lanarkshire data from January to March 2018.

West Lothian have indicated that the expenditure for 2013/14 is an underestimate. This appears to have been due to issues in setting up the IT system for this data collection at the start of the scheme in 2013. Information about expenditure was collected manually rather than through ProcXed and cannot now be updated. West Lothian have indicated that this has led to an ongoing underspend being calculated for the local authority. Additionally, other local authorities have indicated that the figures published for annual over- and underspend do not always match local authorities' own accounts.

Where Scottish Government receive updated information about expenditure from local authorities, the statistics for previous quarters are updated to reflect this. However, local authorities have a cut off point after the end of a financial year at which point they have to produce accounts. At this point, expenditure for the year is calculated, and over- or underspend is calculated. At this point the council meets the overspend, or rolls forward the underspend to the next financial year's budget. If there is a change in the value of an award after this date, this will not be reflected in local authority calculation of the available budget for the next financial year. Additionally, some local authorities may choose to top up the budget for the fund at the start of a financial year with additional funds.

In previous publications, we have included funds provided by local authorities in the available budget, however for this version of the publication this funding has been removed from calculations. Available budget therefore only represents the amounts allocated by Scottish Government plus any underspend from previous years, and it is assumed that local authorities meet any overspend each year.

Each year the discrepencies between these annual expenditure figures and local authorities' accounts are compounded. We are reviewing these differences and to realign our calculation of local authorities' over- or underspend.

Joint Community Care Grant and Crisis Grant applications

Households can apply for both Community Care Grants and Crisis Grants in the same application. In versions of the publication covering the period up to December 2017, total applications reported were the total number of applications recorded on local authority IT systems. In some local authorities, these applications in some cases will include both a Community Care Grant and a Crisis Grant element. However, some local authorities now have a facility to split a single application into the Community Care Grant and Crisis Grant elements, and treat these as two separate applications. Where local authorities have used this facility on their IT system, Scottish Government would receive this as two separate applications, with no information about whether they had originally been made together as one application. Because of this, we have changed the methodology for how total applications to Scottish Welfare Fund are reported, from the 2017/18 publication onwards. For all local authorities, we now report the number of Community Care Grants and number of Crisis Grant applications added together. Therefore, joint applications for Community Care Grants and Crisis Grants will now be counted in the figures as two applications. This makes the figures for total Scottish Welfare Fund applications consistent across all the local authorities.

In addition to the issue described above, local authorities have informed us that on some occasions, applicants may tick boxes on their application forms to apply for both a Community Care Grant and a Crisis Grant, however, based on the types of item, which have been requested the local authority, will process this application as only a Community Care Grant (e.g. if only flooring requested) or only a Crisis Grant (e.g. if only food requested). In these cases, the statistics in this publication reflect the way in which an application has been processed by the local authority, rather than the original tick boxes selected by an applicant.

Payment Methods

An error in data processing has been discovered in regard to the Payment Method categories of 'Cash alternative' and 'Payment into bank'. This error has now been fixed and the time series has been revised to reflect this update.

Reviews

Orkney are unable to supply Tier 1 review information via the ProcXed system. Orkney have supplied full details about all reviews carried out separately and these numbers are included in the tables, however as numbers are small these values have been supressed for disclosure control.

East Ayrshire were unable to submit data on Tier 1 reviews via the ProcXed system until July 2018, but were able to supply summary information about numbers of review applications and decisions separately for 2014 onwards, which is included in the publication tables. From July 2018 onwards East Ayrshire have supplied full review information via ProcXed.

Improvements

We are working with local authorities to make the data more consistent across data fields and Local Authorities. We are reviewing how local authorities interpret the guidance and record information and also identifying issues that are due differences in IT systems.

Management information

Potential data quality issues in the monthly management information are described in our recent pubilcations available at the Scottish Government monthly management information web page.

Comparisons with other UK Statistics

Annual reports by the Secretary of State for Work and Pensions on the Social Fund for the period prior to 1 April 2013 are available from the Department for Work and Pensions. The most recently available report can be found at the DWP Social Fund annual report 2012-13 web page.

On 1 April 2013, the Department for Work and Pensions devolved the responsibility for Community Care Grants and Crisis Loans to the Scottish Government, the Welsh Government and directly to English local authorities.

As a single national scheme covering the UK is no longer in operation, comparable statistics for other parts of the UK are not available.

Boxplots

Boxplots have been included to show the variation in processing times for applications within each local authority (Chart 4b). For each application received to the Scottish Welfare Fund, processing time has been calculated based on the number of working days between the date all information was received and the date the initial decision was made. A processing time of zero days indicates the application was processed on the same day that all information was received. The boxplots show (i) the average value (this is the median, or middle ranked value) of processing times across all applications within a particular local authority, and (ii) the variation in processing times among the applications within a local authority.

The variation among processing times is shown using a 'box' and lines extending out from the box (illustrated in the example below). The box shows the range of values around the median within which half of the processing times from a particular local authority fell. The lines extend out from the box to show the range of values within which 90% of the processing times from a particular local authority fell. The larger the box, and the longer the lines, the more variation there is among the processing times within the local authority.

In the example above:

The median value for processing times was ten days. When all the applications' processing times were ranked in order, the middle value was ten days. Half of the processing times fell below this and half above this.

The width of the box was relatively small, indicating that half of the applications were processed in five to 15 days.

However, some applications took much longer to process. The end of the line to the right of the chart indicates that 95% of applications took up to 30 days to process.