As an outsider to the US criminal justice system, it has been fascinating to witness all of the activity in the United States directed at pre-trial reform.

Driven largely by the costs of mass incarceration, jurisdictions are looking to reduce their pretrial detention rates as one way to address jail overcrowding. In the US, like NZ, much of the growth in the incarcerated population has been in the pretrial population.

Recent analysis shows that 62.8% of the jail population are people who are awaiting trial (this represents around 450,000 people at any one time, and close to 30% of the total incarcerated population). This number has almost doubled in the last 15 years. New Zealand’s pretrial population has also doubled in the last 15 years, and now represents close to 20% of the total incarcerated population.

A short note on comparing the US and NZ criminal justice systems. Despite their common origins, the NZ and US criminal justice systems operate somewhat differently. Even our terminology is different. For example, the word “jail” is commonly used in the US to describe a pre-trial and short sentence incarceration facility (often run by counties), whereas “prison” is usually reserved for people serving sentences of beyond 1 year (usually run by the State). In NZ, the terms “jail” and “prison” are used interchangeably, and NZ’s pretrial prisoners are held in high security facilities (usually part of a prison) and are mixed with those awaiting sentence. We describe these facilities and prisoners as “remand”.

In addition, my observation is that there is generally less discretion at the front end of the system in US jurisdictions than in NZ (which operates a unified system). In many US jurisdictions, Police often operate under mandatory arrest policies, even for quite minor offences (of course this differs state to state), and charging decisions are the responsibility of District Attorneys. In NZ, Police have greater discretion as to their arrest and charging decisions, and a wider range of options to divert people away from the criminal justice system through a series of warnings, formal warnings, adult pre-trial diversion, and (in some locations) referral to community justice panels.

This means that the number of people initially detained pre-trial in the US is higher than in NZ. This also means that it is not overly meaningful to compare our pre-trial release rates, as the US is likely to have a higher arrest and detention rate to start with, as well as differences in what is considered criminal/not. And finally, NZ does not use any form of cash bail or secured money bonds within its bail system, so the decision-making incentives and considerations are different.

The National Institute of Corrections has been a longstanding leader in what is now a crowded field of bail reform organisations and advocates. This was acknowledged by Judge Truman Morrison at the beginning of the weeklong Orientation for Pretrial Executives, held in Aurora, Colorado in early March. In particular, the Judge acknowledged the NIC’s intellectual leadership, forethought and fidelity of commitment to pretrial reform, and branded the NIC as the forerunner in pretrial reform.

The pre-trial training took a holistic approach to bail reform, in that it focused on all the necessary elements of a high-functioning pretrial justice system. This framework comprises:

• the law
• standards and essential elements of pretrial systems and agencies
• research and evidence-based practices
• organisational theory.

If there was one thing that I took away from this training, it was the message that bail reform could not be achieved without focusing on all four of these elements: a technocratic solution alone is not sufficient, law change alone is not sufficient, organisational and practice changes are not sufficient.

What does the research tell us?

Recent research into the outcomes for those detained pretrial has been instrumental in demonstrating the negative – and ongoing – impacts of pretrial incarceration. Using matched US datasets, researchers have shown that:

• those without legal representation at their bail hearing are more likely to be remanded in custody
• those held in custody pre-trial are more likely (than their matched counterparts) to:
  • plead guilty
  • be convicted
  • receive longer sentences
  • have worse reoffending outcomes (cite)
• defendants detained for their entire pre-trial period are over four times more likely to be sentenced to a term of imprisonment than defendants released at some point, with the results being even more pronounced for low risk defendants (cite)
• that even more than 24 hours in custody for low or moderate risk defendants makes them more likely to reoffend than matched counterparts (cite).

Judge Truman Morrison expressed his dismay: “No experienced judge in America was aware of these numbers and the extent to which we shoot ourselves in the foot by keeping low and medium risk people in jail for even one day.”

My particular interest in the training was to learn about the use of pretrial risk assessment, an area in which the US is world-leading in terms of the development and use of pretrial risk assessment tools. Before I turn my focus to what I learned about pretrial risk assessment, a short note about the course itself. It is an intense 4-day course (hence the “bootcamp” moniker coined by some of the other participants), covering everything one needs to know about pretrial – the history of bail (right back to the Norman Invasion), the legal foundations of bail, the goals of pretrial justice, the essential elements of a pretrial system and agency, legal and evidence-based practices to ensure pretrial safety and appearance, and organisational theory and culture change. The faculty were all experts in their field and very engaging presenters.

They included Judge Truman Morrison, senior judge in the DC Superior Court, leadership and organisational theorist Tom O’Connor, legal expert Tim Schnacke, Kentucky Pretrial Services Executive Tara Boh Blair, pretrial and community corrections expert Don Trapp, and a range of local practitioners who shared their experiences of pretrial reform. The aim of the course is to equip pretrial practitioners with the tools and knowledge they need to undertake pretrial reform in their jurisdictions – and given the energy put into action planning that occurred on day 4, I was left with little doubt that it achieved this.

Pretrial risk assessment

Validated and well-implemented risk assessment is regarded as one of the essential elements of a high-functioning pretrial system.

As one of the facilitators put it, “we’ve been doing risk assessment since 400AD”. Certainly in NZ, there is a risk assessment made by the judge on the basis of information provided by the prosecution and defence. However, the data-driven tools now used in many jurisdictions in the US have been shown to be better predictors of outcome than decisions made without validated risk assessment.

It is important to remember that a risk assessment tool is only as good as:

• the data used to create it – many States have developed their own tools, because some factors are not predictive of risk in some States, often due to different methods of data capture or different offender populations (for example, Colorado found during the development of its tool that previous failures to appear were not predictive of future risk of non-appearance )
• its implementation – there is a need for widespread discussion and agreement on what risk is appropriate, why the assessment is being implemented and how it works (presenters gave examples of judges systematically ignoring risk assessments, and in one State, an example of implementation failure where Police did not buy in to the use of the tool and started stacking charges to make a defendant appear more risky)
• its retesting and validation, as well as quality assurance across those administering it (for example, DC recently removed marijuana use from its risk assessment tool – not because it became legal – but because their re-validation showed that marijuana use was not predictive of failure to appear or reoffending).

A risk assessment tool does not:

• tell you that individual’s risk (it just tells you that the person has the same characteristics as a group of people with a particular risk profile, and their likelihood of success on bail)
• remove judicial discretion – it simply informs the decision
• tell a decision-maker what to do!

The “what to do” is the hard part. A risk assessment tool may need to be coupled with guidance on the appropriate response to ensure consistency of implementation. Where risk assessment tools have been introduced, there has often been agreement about what the scores mean, supported with tools to guide decision-making.

There is plenty of knowledge and experience here in the US for NZ to build upon if it chooses to go down the path of incorporating more formal risk assessment into pretrial decision-making. That said, a significant amount of work would be needed to lay the groundwork for such a tool, not least of which is the fundamental research that has already been completed in the US.

However, my initial view is that if the development of a risk assessment tool for the NZ population is feasible, such a tool, if properly implemented and validated with broad discussion on what the tool is and what the scores mean and for whom, would improve consistency of bail decisions, improve confidence in bail decisions, and provide a more rigorous basis for conversations about pretrial risk and risk acceptability in NZ. Moreover, there is the possibility that incorporating more formal risk assessment into pretrial decision-making could improve both individual and system outcomes as has been the case in the US – and isn’t this what we’re all working for?