Abstract:
Belgium is one of the last countries in Western Europe where no national recidivism statistics are available. As a consequence, criminal policy can only make use of international statistics and of studies of specific, often unrepresentative groups of delinquents. In this study, use is made of the Central Criminal Registry (CCR). All persons with a CCR registration of a court-imposed measure or sentence in 1995 (n = 136,530) are checked for new court-imposed measures or sentences in the CCR (‘recidivism’). 49.3% of the cohort had no court-imposed measure or sentence prior to the registration in 1995 (‘antecedent’). The recidivism of this cohort is studied until November 2013. During the entire follow-up period, 57.6% of the cohort has recidivated (prevalence of recidivism). 50% of the recidivists had a new conviction after at most 742 days, barely over 2 years (duration to recidivism). 70% of all recidivists have at least 2 new convictions (frequency of recidivism). Taken together, the recidivists in the study have received a total of 405,781 new convictions. A survival analysis has been conducted. The results show, inter alia, that people who are younger in 1995 have an increased risk to recidivate. The risk of recidivism is higher when people have more antecedents. Women have less risk to recidivate than do men. Offenders convicted to a prison sentence have an increased risk of recidivism in comparison with other offenders. The results of the study provide gross recidivism base rates. Limitations of the study are related to the limitations of the data base, including the lack in information about the application of sentences. This descriptive study is a first step in the direction of developing in-depth recidivism research based on large samples of offenders. Future research based on this study could focus on criminal careers, mapping of the effects of specific penal interventions and the relations of offending with a range of life domains.