Written evidence submitted by the Centre for Public Data (REM0027)

1.  Executive summary

●        We recommend that the Ministry of Justice start publishing data on the length of time people spend in remand and the reasons why remand is used. This evidence could help reduce the unnecessary use of remand and reduce the remand population.

●        Our research suggests this data is recorded internally and should be straightforward to extract from HMCTS and HMPPS databases.

●        The remand population is at its highest point in decades, putting strain on the prison and court systems. Yet basic data is missing on the remand population and why remand is used, which affects our ability to reduce the unnecessary use of remand.

●        Key data gaps include:

a.      Length of remand: No routine statistics are published on how long people are remanded in custody, which makes it hard for researchers to track the population, and we don’t know e.g. the number of defendants held longer than the maximum sentence for their offence.

b.      Reasons for remand: No official data is published on the legal grounds on which bail is refused, so researchers do not know whether remand is being used unnecessarily.

●        Our research suggests that information on the above is already held on prison and court databases in structured form, and would be straightforward to publish:

a.      Length of remand: HMPPS’s prisoner management system records information on remand prisoners including dates and offence type, and MoJ extracts data from this routinely - it should be straightforward to add statistics on the length of time that defendants have been held in remand.

b.      Reasons for remand: HMCTS’s court systems records the reasons why a court refuses remand, as do police forms, in structured form. MoJ can

extend its existing statistical publications to include information on the reasons why remand is refused, broken down by offence type.

2.    Data gaps on remand

We identified two areas commonly raised by justice system stakeholders, where better data can help ensure that remand is only used where warranted. Our research suggests that resolving these data gaps should be straightforward, since most of the requisite information is stored in structured digital form. It should therefore be straightforward to extract and publish it.

2.1  Length of time people are held on remand

As of June 2022, 13,409 defendants were held in remand, 27% of whom had been held for more than 6 months.[1] This information was obtained via a Parliamentary question, as the MoJ does not routinely publish statistics on the length of time people are held on remand though this appears to be the key driver of the recent rise in the remand population.

This lack of data makes it harder to analyse the reasons for the increase in the remand population and how it can be addressed. Charlie Taylor, HM Chief Inspector of Prisons, told MPs that “One of the bits of data we would be very keen to see published would be for how long people had been remanded. We know what the remand population is at any one particular time, but what we do not know is how long people are being remanded for and the extent to which that is changing.”[2] Many other justice system stakeholders have called for this data to be published routinely.[3]

2.1.1 Available data and recommendations

Data on the length of time spent in remand is not routinely published, but MoJ supplies ad hoc data.[4] This is drawn from Prison-NOMIS, an operational database used in prisons for the management of offenders.[5] Given the number of FOIs the MoJ receives on remand lengths it would be more efficient to simply add it to routine publication.

We have established that Prison-NOMIS records the following fields per defendant[6]:

●        The ‘reception date’ on which the defendant entered custody

●        The custody type - remanded on bail or sentenced

●        The offence committed

●        Other information on prisoners, including personal details and age.

It should thus be straightforward to extract data from Prison-NOMIS routinely, including any defendants held beyond the maximum sentence for the offence with which they are charged. We recommend:

1.      As recommended by the Prison Reform Trust and others, the average length of time that defendants are held in custody, plus the number of defendants held beyond 6 months, 1 year and 2 years, should be routinely published to aid scrutiny.[7]

2.      This should include offence type (to identify defendants being held for long periods for non-violent offences) and count of defendants held beyond the maximum sentence awarded for their offence (to identify unwarranted use of remand), and should be disaggregated by ethnicity, age and sex, to help understand disparities in the use of remand.

3.      This can be implemented straightforwardly by MoJ statistics teams working from Prison-NOMIS, who already receive a regular extract from Prison-NOMIS to produce the Offender Management Statistics Quarterly.[8]

2.2    Reasons why bail is refused

When considering remand, the police and courts rely on the Bail Act 1976, the legal framework for when bail can be refused. These reasons include risks that the defendant would not surrender to bail, commit further offences, or for the defendant’s own protection.[9]

The Government does not collect or publish data on the reasons courts refuse bail - data which many justice stakeholders have called for.[10] Without this data, as Dr Tom Smith from the University of West England has argued, it is difficult to understand fluctuations in the rate of remand and why its use has increased in recent years.[11]

In addition, many stakeholders have expressed concern about the use of remand for a defendant’s ‘own protection’. Both the Justice Select Committee and an APPG have recommended that the Government bring forward legislation to abolish this power[12], but the Government has not been able to supply MPs with data on the use of this power.[13]

2.2.1  Available data and recommendations

It should be feasible to publish data on why bail is refused, as our research suggests that data on this is already stored at least two points within the CJS:

●        Police recommendations to refuse bail: Firstly, when the police make an initial recommendation to refuse bail, MG7 forms include a tick-box for the reason for the decision.[14] These forms are sent in digital form to the CPS. Given this structured digital recording, it should be possible for either the Home Office or the CPS to publish aggregated data on the reasons why police recommend refusing bail.

●        Court decisions to refuse bail: Secondly, when a case reaches court, legal practitioners have told us that Common Platform requires court clerks to record the reason why a court has refused bail, from a drop-down menu. A drop-down menu in a digital system implies an underlying structured database from which information could be easily extracted. Common Platform is now used in 60% of courts, which would provide an adequate basis for management information.

We recommend that:

1.      The Home Office consider updating the Annual Data Return supplied by police forces[15] to include data on why police make the recommendation to refuse bail, and include this information in its Police Powers and Procedures statistics.[16]

2.      MoJ extracts data on the reasons supplied for remand decisions from Common Platform into the Court Proceedings Database (the source for its Criminal Justice Statistics Quarterly publication, including its remand pivot tables), and based on this, publish management information (or augment existing remand tables within CJSQ) to provide information on the reasons given for refusing bail - this would also allow analysis of decisions broken down by offence type and defendant characteristics.

This can be done immediately and published as management information, with suitable caveats - there is no need to wait for Common Platform to reach 100% of courts.

December 2022