Relative Impact of Underreporting and Desistance on the Dark Figure of Sexual Recidivism

Abstract

Sexual recidivism rates based on arrests or convictions underestimate actual reoffending due to underreporting. A previous Monte Carlo simulation estimated actual recidivism rates under various reporting and conviction assumptions but did not account for desistance—the decreasing likelihood of reoffending over time. This study addresses that gap by incorporating a 12.3% annual desistance rate (from a well-known empirical study) and exploring its impact alongside varying charge rates (100%–5%). The results showed that reductions in charge rates lead to disproportionately large increases in recidivism. For instance, lowering the charge rate from 50% to 25% results in a much larger increase in actual recidivism than reducing it from 100% to 75%, despite both being 25% reductions. This indicates that as charge rates decrease, actual recidivism grows more sharply. A sensitivity analysis also examined desistance rates of 0%, 5%, 12.3%, and 20%. Higher desistance rates cause reoffending to occur earlier but have little impact on long-term totals. Over 25 years, reoffending rates remain similar across desistance rates, suggesting desistance affects the timing, but not the overall amount of reoffending.