Association for Public Policy Analysis & Management

Introduction: Whether or not it is sensible to base the penalties for drug distribution on the weight of the drug involved, that is in fact the principal method used in the USA, in particular as reflected in the Guidelines of the US Sentencing Commission. Scholars have debated the role and functions of sentencing, even for drug offenses, commenting critically on their extreme severity. Yet, little has been written on how sentences are set relative to the level of harms arising from the drug in question. This paper assesses whether it is possible to provide a credible or authoritative basis for this kind of harm ranking across drugs with available data.

Method: After discussing the functions of sentences and how they apply to drug offenses, we then evaluate drug sentencing guidelines for four drugs (crack cocaine, methamphetamine, MDMA, and BZP). We examine how the Commission measures harms and determines sentence structure, giving emphasis to the problem arising from the limited data available when guidelines are initially developed. This is followed by a review of the existing literature that aims to provide an index of drug harms, showing that it is both conceptually and empirically flawed. Finally we consider an alternative approach that uses population level data on harms (e.g. overdose deaths, number of problematic users) and quantities (kilograms consumed) to measure harms per gram. These measures of harms are used to compare crack cocaine and methamphetamine. In order to account for the uncertainty of key parameters, we employ a simulation model.

Results: Preliminary analysis shows, more harm per gram of methamphetamine consumed for three of the four population-level indicators considered. When we assess the harms associated with the minimum quantity threshold that trips a five year sentence under the guidelines (28 grams of crack; 5 grams of methamphetamine) we find 28 grams of crack is responsible for more harms than 5 grams of methamphetamine on all indicators and generally by a wide ratio. The final paper will include a number of additional comparisons.

Discussion: Our preliminary findings suggest that this alternative assessment of harm might be more accurate in gauging harms, which could lead to more procedurally just sentencing but also point to the fragility of any effort to make such comparisons for sentencing purposes.