Author Evarn J. Ooi
Published September 2020
Report Type Crime and Justice Bulletin No. 228
Subject Recidivism / Re-offending; Parole; Evaluation reports
Keywords Practice Guide for Intervention (PGI), risk-need-responsivity (RNR), cognitive behavioural therapy (CBT), recidivism, parole



We investigate the impact of the Practice Guide for Intervention (PGI) on re-offending among high-risk parolees in New South Wales (NSW).


Introduced in June 2016, PGI was a major component of the ‘Enhanced Community Supervision’ reform and led to a dramatic overhaul in the delivery of supervision services. Using a difference-in-differences (DiD) strategy, we compare re-offending behaviour between offenders released from prison on parole and those released unconditionally before and after the introduction of PGI. PGI is compulsory for offenders released on parole with a Level of Service Inventory-Revised (LSI-R) score of medium or above, and consequently, the sample is limited to offenders with these LSI-R scores. Re-offending is measured as the probability of committing a new and proven offence within 12 months of release from prison. The pre-PGI period includes offenders released from prison between June and December 2014. There are two post-PGI periods. The first post-PGI period includes offenders released between June and December 2016, which coincides with the first six months after PGI was introduced in NSW. The second post-PGI period includes offenders released between June and December 2017, when the use of PGI across NSW was approaching its historical peak.


A comparison of the trends in the re-offending rate before the introduction of PGI confirms that prisoners released unconditionally form a natural comparison group for parolees. The DiD estimates reveal a 2 to 3 percentage point reduction in the likelihood of re-offending among parolees compared with those released unconditionally after the introduction of PGI. However, the estimates are not statistically significant.


The results suggest that the introduction of PGI did not have a statistically significant impact on re-offending rates of high-risk parolees.