Occasional paper: The Effect of COVID-19 on Community Payback Order Unpaid Work or Other Activity Requirements

Annex A

Legislation

More information on The Community Orders (Coronavirus) (Scotland) Regulations 2021 is available in the Legislation.gov.uk website.

Data sources

There are three different sources of data used in this analysis:

1. Unit level data up to year 2021-22 on community payback orders, provided to the Scottish Government by local authorities for inclusion in the annual justice social work statistics publications.

2. Unit level data provided from the Scottish Government disposals dashboard, which enables this model to estimate for 2022-23.

3. Aggregate management information collected quarterly on the number of unpaid work requirements and hours outstanding.

A spreadsheet has been published alongside this paper with the model estimates and management information included.

Throughout the paper the terms “start date” and “end date” are used. These relate, respectively, to the dates on which the unpaid work or other activity requirement was imposed and the date on which it was completed/terminated.

All of the statistics were sourced from Justice Social Work Statistics unless otherwise stated.

Stage 1: Successfully completed requirements (building block 1)

Successfully completed includes all unpaid work requirements that were successfully completed or discharged early.

There were 33,143 such requirements between 2017-18 and 2021-22. For these requirements, information is held for:

• Date requirement imposed,
• Date requirement completed,
• Hours originally imposed,
• Hours of both unpaid work and other activity which were done.

This analysis has split these records down into different sub-categories but there was one common general assumption made for each of them. This was that the hours done were evenly spread over the life of the requirement.

These 33,143 records were analysed by the following sub-categories:

1. The biggest category (27,570 records) was for requirements where the number of hours done was exactly equal to the number of hours imposed. The estimates for hours outstanding at the different points was estimated by assuming even spread over the lifetime of the requirements.

2. There were 2,472 records where either:

a. More hours were done than were imposed, or

b. Less hours were done than were imposed but they were within 5 hours of each other.

These were treated in the same way as the first sub-category, with the hours outstanding based on the hours imposed rather than the hours done.

3. Requirements where information on hours completed was available but hours imposed was not (60 records). The hours imposed was assumed to be the same as hours completed for these records and they were subsequently calculated the same as the first sub-category.

4. There were 262 records where:

a. Less hours were done than were imposed and the difference was more than 5 hours, and

b. The requirement finished before the first national Covid-19 lockdown began on 26 March 2020.

This was made a separate sub-category as the number of hours imposed in some orders was reduced , but that was not the case with these records. The hours outstanding were also calculated based on the hours imposed.

5. Similarly, there were 94 records where:

a. Less hours were done than were imposed and the difference was more than 5 hours, and

b. The requirement finished after the first national Covid-19 lockdown began on 26 March 2020 but before 15 March 2021 when the number of hours imposed in a large number of orders was reduced by 35%.

For these 94 records, the hours outstanding were also calculated based on the hours imposed.

6. There were a further 68 records where:

a. Less hours were done than were imposed and the difference was more than 5 hours, and

b. The requirement was imposed after 15 March 2021 and so was not affected by the 35% reduction in the hours imposed.

For these 94 records, the hours outstanding were also calculated based on the hours imposed.

7. There were always going to be a large number of records which were in existence on 15 March 2021, and which were therefore going to be affected by the 35% reduction in hours imposed. There were 2,511 requirements where:

a. They finished on or after 15 March 2021, and

b. The hours not done formed no more than 38% of the total hours imposed. 38% was used as opposed to 35% in order to allow for rounding of hours.

A similar calculation was made based on hours imposed, although there was one slight difference. On dates prior to 15 March 2021, the hours outstanding during the life of the requirement could not fall below the number of hours which were taken off by 35% reduction in hours.

For instance, suppose a person was given 120 hours on 9 April 2020 and, following the reduction in hours imposed by 35% (42) on 15 March 2021, they successfully finished the requirement on 3^rd April 2021. The hours outstanding were assumed to be done consistently over time. In this case that meant assuming the full number of hours being done in 360 days – an average of 0.333 hours per day. The outstanding hours at various points for such an order would be:

• By 5^th October 2020 – 60 hours (120 less 60 hours done in 180 days)
• By 4^th November 2020 – 50 hours (60 less 10 hours done in 30 days)
• By 4^th December 2020 – 40 hours (50 less 10 hours done in 31 days)
• As at any date from 5^th December 2020 to 14^th March 2021 inclusive – still 40 hours
• As at any date from 15 March to 2 April 2021 inclusive – 0 hours as the 35% reduction had been applied by that point (ie. 40 hours is less than 35% of 120 (42 hours))
• By 3^rd April 2021 – 0 hours as that is the date on which the requirement finished.

8. The remaining 106 records were ones where:

a. They finished on or after 15 March 2021, and

b. The hours not done formed more than 38% of the total hours imposed.

These requirements were assumed for the model to have not been affected by the 35% reduction in hours imposed. Therefore, the calculation of hours outstanding was a straight calculation based on hours imposed with no adjustment.

Stage 1: Requirements which finished without being successfully completed (building block 2)

There were 17,050 such requirements between 2017-18 and 2021-22. This covered requirements which did not reach a successful completion, for various reasons such as the order being revoked, the order being transferred out of the area or the death of the individual. For these requirements, information is held for:

• Date requirement imposed,
• Date requirement terminated,
• Hours originally imposed.

No information is available from the unit level collection for such requirements on the hours of unpaid work or other activity which were actually done. Therefore, while it is known that there were no hours outstanding following the requirement being terminated, the hours outstanding at points during the life of the requirement needs to be estimated.

An assumption has been made that all unpaid work hours that have not been done remain outstanding until the actual date on which the requirement is regarded as formally terminated.

In this analysis, calculations were made using different limits for the proportion of hours which were actually done for these requirements but the upper limit range is calculated using 0% and the lower limit is 100%.

As with requirements that were successfully completed, the assumption was made that hours done were evenly spread over time (ie a daily rate)..

Stage 2: Requirements still in progress at end of March 2022

There were 7,966 such requirements over the five years. For these requirements, information is held for:

• Date requirement imposed,
• Hours originally imposed.

The exception was 5 records which did not have information on the number of hours imposed. This exercise therefore proceeded with 7,961 records.

The biggest issue with these 7,961 records is that it is not known whether they have been successfully completed. Clearly though, some will successfully complete in the future, and some will not.

To get an idea of the likely outcome of these requirements, it was possible to go back to earlier years’ data. For CPOs outstanding on 31 March 2018 which had an unpaid work requirement, 65 per cent of these where the outcome was known had gone on to be successfully completed. For those outstanding on 31 March 2019, exactly the same proportion were successfully completed.

It was assumed that 65 per cent of those outstanding at end March 2022 would go on to successfully complete. An exercise was therefore conducted to randomly allocate the 7,961 records as follows:

1. 5,175 requirements were assumed to go on to successful completion, and

2. 2,786 requirements were assumed to terminate without successful completion.

Stage 2: Requirements in progress, assumed to successfully complete (building block 3)

For those assumed to successfully complete, it was necessary to estimate the date on which these requirements would finish. It is known that the time it takes to finish a requirement varies by:

1. The number of hours imposed, and

2. The year the requirement was imposed. For example, requirements which began during the years that were most affected by Covid-19, generally took longer to complete.

The following table shows, for successfully completed unpaid work requirements which finished in different years, the number of days it took for them to complete. (Please note, comparison to Justice social work statistics is not exact as this information was created from the data used within the model).

These time periods were taken from the unit level data on CPOs collected from local authorities and were the average time taken for orders which completed successfully in those years. The longer time periods for 2020-21 and 2021-22 clearly reflect the pandemic, in particular the fact that the Coronavirus (Scotland) Act 2020 extended the time limit for the completion of existing unpaid work requirements in CPOs by 12 months, and required that any new requirements imposed during the period in which the legislation was in force had a time limit of at least 12 months . This came into effect on 7 April 2020. This resulted in particularly large increases in the time taken to complete level 1 (100 hours or less) requirements.

For each of the years 2017-18 to 2020-21, the time taken for requirements which finished in those years was felt to be a good proxy for the time taken for ones which began in those years. However, for 2021-22, as fewer and fewer restrictions were in place as that year went on, it was felt that the best figures to use were the average of the time periods for requirements finishing in that year and ones finishing in the most recent pre-pandemic year of 2019-20.

Where these time periods resulted in the calculation of a date of completion that was on or before 31 March 2022, the predicted completion date was defaulted to 1 April 2022 as it was known that the requirements were in existence at close on 31 March 2022. This affected 1,547 (30 per cent) of the 5,175 records. This relatively high proportion undoubtedly reflected the much longer time it was taking because of the Covid-19 pandemic to get orders to a conclusion.

Once the date of assumed successful completion was calculated, the calculations for hours outstanding were based around hours imposed and assumed to be evenly spread during the (assumed) life of the requirements.

Stage 2: Requirements in progress, assumed to terminate without successful completion (building block 4)

For those assumed to not successfully complete, it was necessary to estimate the date on which these requirements would terminate. Again, it is known that the time it takes to finish a requirement varies by the number of hours imposed and the year the requirement was imposed. (Please note, comparison to Justice social work statistics is not exact as this information was created from the data used within the model).

The following table shows the number of days it was assumed that each requirement would take to terminate:

As per building block 3, these time periods were taken from the unit level data on CPOs collected from local authorities and were the average time taken for orders which terminated unsuccessfully in those years. They were used as a proxy for the time taken for ones which began in each of the years 2017-18 to 2020-21. Again, the exception to this was for 2021-22, where it was taken as the average of the time for requirements terminating in that year and ones terminating in the most recent pre-pandemic year of 2019-20.

Where this resulted in a date of termination that was on or before 31 March 2022, the date was changed to 1 April 2022 as it was known that the requirements were in existence at close on 31 March 2022.

Once the date of assumed termination was calculated, the estimations can be made for hours outstanding. Again, as with requirements which had been terminated without success, it is not known what proportion of hours would go on to be completed before termination. Therefore, calculations needed to be done using different values for this proportion.

The calculations were based around hours imposed and hours that were assumed to be completed before termination. They also assumed hours were evenly spread during the (assumed) life of the requirements.

Final Stage: Requirements which were imposed between 1 April 2022 and 31 March 2023 (building blocks 5 and 6)

Unit level data on CPOs for year 2022-23 is being collected during 2023 and will not be available for use until it is published. This is likely to be in early 2024. Data supplied by SCTS to JAS for use in the disposals dashboard shows charges that are disposed with a CPO with an unpaid work requirement attached. This data is available for 2022-2023.

The management information used in the disposals dashboard shows the data at charge level, whereas the local authority data is at CPO level. For this analysis and, to adjust it to the local authority data, only the maximum number of unpaid work hours per case-accused (for all the charges resulting in a CPO that the accused can have) in each month have been extracted from the disposals dashboard management information.

The disposals data for this period (relating to CPOs with unpaid work requirements) contained 9,362 records and, for these requirements, information is held for:

• Date requirement imposed,
• Hours originally imposed.

Therefore, like the orders in Stage 2, it was necessary to conduct an exercise to:

1. Randomly allocate the 9,632 records between those assumed to go on to successful completion and those assumed to terminate without successful completion, again assuming that 65% of the orders would go on to successfully complete, and

2. Estimate the length of time the order would take to complete/terminate.

The 9,632 records were therefore randomly allocated as follows:

1. 6,261 requirements were assumed to go on to successful completion, and

2. 3,371 requirements were assumed to terminate without successful completion.

No information is currently available on the length of time requirements imposed from April 2022 onwards have been taking to finish. The following table shows the number of days it was assumed that these requirements would take to complete/terminate. This has been based on the assumption that the periods for year 2022-23 are now similar to pre-pandemic and the figures in this table are the average of the figures for the three pre-pandemic years 2017-18 to 2019-20.

Once the date of assumed completion/termination was calculated, the estimations are made in the same way for hours outstanding. Again, as with requirements which had been terminated without success, it is not known what proportion of hours would go on to be completed before termination. Therefore, calculations needed to also be done using different values for this proportion.

Again, the calculations were based around hours imposed and hours that were assumed to be completed before termination. They also assumed hours were evenly spread during the (assumed) life of the requirements.