Justice data gaps: How to improve data on conditional bail
This post is the latest in a series from our project to map and fill data gaps in criminal justice.
Let’s imagine that the worst has happened, and that you have been arrested, taken to the police cells, and charged with a crime. Your court date isn’t for a few weeks. What happens to you while you await the chance to prove that you’re innocent?
There are three possibilities. Option 1 is that you’re trusted to go home until you appear in court - this is called ‘unconditional bail’. Option 2 is that you’re considered so dangerous that you’re locked up in prison - this is called ‘custodial remand’. Option 3 is ‘conditional bail’.
Conditional bail means you are free to go home until your day in court, but have to follow restrictions. These could include curfews, reporting to police, or even an electronic tag.
At its heart, conditional bail is a mechanism to juggle the different goals of the criminal justice system: protecting the public, upholding civil liberties, reducing the pressure on prisons, and making sure defendants turn up at hearings.
With the prison remand population at its highest in 50 years, it's easy to see why the police and courts are increasingly keen on conditional bail - experts say that conditions are fast becoming the norm.
Why are we looking at it?
Although tens of thousands of people are now subject to conditional bail each year, we’ve learned (via our research into justice data gaps) that virtually no data is published about it.
Even the most basic facts are unknown - how many people are on conditional bail, which conditions are used, or how long they last.
This is alarming, because conditional bail involves restricting the liberties of people who aren’t convicted of any crime, yet has little oversight. Professor Anthea Hucklesby warns that the “use of conditions is completely unregulated” and that courts can “attach whatever conditions they want for however long”.
Additionally, poor data means we just don’t know whether conditional bail works, even though it costs the police money to use. The Criminal Justice Alliance warn of a lack of trust in conditional bail due to a lack of monitoring. And there is no clear evidence whether conditional is better than unconditional bail at preventing offending.
In short - we don’t know how much conditions are being imposed, how much they’re costing, or whether they’re actually keeping the public safe.
What are the main data gaps?
Our research, and interviews with experts, suggested that the most basic data gaps are as follows:
How many people receive conditional bail, and who are they?
What conditions are imposed, and for how long?
Does conditional bail achieve its aims?
We set out to establish whether these could be answered, by digging into out bail data is recorded, where it goes, and how it is published (see our full briefing for the full findings).
The positive news is that data on conditional bail is recorded at multiple points across the criminal justice system. As this data is recorded digitally, it should be straightforward to extract. Based on this, we’ve come up with three things we think the Ministry of Justice can do to help.
Recommendation 1: Add conditional bail to current bail statistics
Current MoJ court bail statistics don’t distinguish between unconditional and conditional bail - but we think they could. We have been told that Common Platform, the case management system for criminal courts, does record information on conditional bail, including the conditions applied.
As Common Platform is a digital system, it should be technically possible for the MoJ to extend its current data extracts to distinguish between court defendants given conditional and unconditional bail (see our briefing on how this might work).
This would radically transform our understanding of conditional bail, revealing how frequently it is used and the characteristics of those who receive it - the statistics on bail already allows data to be broken down by age, sex, ethnicity and geography.
Depending on how the underlying information is recorded, it could also be possible to add details of the conditions applied, and how long they’re imposed.
Recommendation 2: Explore using case files to record more on conditions
Another source of data on conditional bail could be police information. Typically, officers pass case information to prosecutors using Manual of Guidance (MG) forms. Form MG4A (bail grant/variation) includes tick-boxes on bail type and the grounds for imposing it, plus free-text fields on any conditions imposed.
These days, most MG forms are submitted digitally. We recommend that the CPS and/or MoJ explore the possibility of using these digital tick-boxes (in effect structured data) to collect data on how often the police use conditional bail, and the grounds for imposing conditions.
The forthcoming Digital Case File standard, which will support a more structured digital exchange of information between the police and CPS, could create new opportunities too. For example, the DCF could include a structured rather than unstructured list of bail conditions, which could help to tackle a key data gap: which conditions are imposed.
Recommendation 3: Find ways to capture reoffending while on bail
Politicians, justice experts and the public all want to ensure that conditional bail is effective. Perhaps most importantly, we need to know if conditions help to keep the public safe.
Whilst overall offending on bail has fallen since 2010, the data available doesn’t show the rates for those specifically on conditional bail. We believe this is because data on the Police National Computer records whether a person arrested was on bail for an offence, but nothing else. We therefore don’t know how conditional bail affects offending, or which conditions are effective.
There are two ways better data could potentially be collected on offending whilst on conditional bail - we recommend MoJ investigate whether PNC data could be used to generate experimental statistics:
Potentially, Police National Computer data could additionally record whether an individual arrested is on conditional bail, and if so, the conditions imposed: this could then be fed into MoJ reoffending statistics.
Alternatively, the MoJ could investigate the possibility of matching PNC and other defendant-level records (as it already does for its reoffending statistics) to establish the relationship between conditional bail and offending.
The future of conditional bail
In the coming years, pressures on the criminal justice system are likely to increase the use of conditional bail further. Recently, the Justice Select Committee concluded that conditional bail “is a good alternative to remanding defendants to custody”. But the Committee were also concerned that a lack of guidance and consistency in its use undermined its effectiveness.
Conditional bail may well be a good way to tackle the court backlog and ease burdens on prisons. But before expanding its use further, we urge the Government to commit to publishing the basic data on conditional bail. We need this so that the public, police and magistrates can have confidence in its use.
For more details, download our full briefing.
This is the latest in a series of deep dives from our justice data gaps project and is funded by the Justice Lab, an initiative of the Legal Education Foundation, as part of their ongoing programme of research and advocacy to improve the quality and availability of justice system data. If you’re interested in this or any of our other work, please get in touch.
Special thanks to Professor Anthea Hucklesby, Griff Ferris at Fair Trials, Penelope Gibbs from Transform Justice, Dr Tom Smith and others for sharing their expertise with us. Any errors are our own.