There are two additional implications of the uncertainty in forecasts of crime rates and offenders: the periods over which crime forecasts are made should be as short as possible and the forecasts should be updated frequently. Large-scale social systems have elements of complexity or nonlinear dynamics and uncertainty that militate against the accuracy of long-term forecasts. In practical terms, this means that forecasting cones (upper and lower bounds) for enveloping the ranges within which crime is likely to fall with a high probability will grow very rapidly from the base year into the future. To take this into account, the time periods of the forecasts should be relatively short and the forecasts should be revised when new information becomes available. For most police, court, and penal components of the juvenile and the criminal justice systems, this is not particularly problematic, as forecasts typically are necessary only for one- or two-year government budgeting cycles. Only occasionally are projections more than five years into the future required for budgeting or planning purposes.
Official data to track or monitor crimes committed by juveniles and the justice system responses to juvenile offenders are clearly inadequate. They provide, at best, only a crude measure of perpetrators estimated by victims to be under 18 or of the number of arrests for the various crimes of juveniles under 18. The reporting of crimes known to the police and arrest data is voluntary on the part of local police agencies and states. Therefore, published FBI annual crime figures are based on different agencies' and states' reports each year, depending on which agencies and states submitted their data on time. Official data are insufficient for studies to determine whether changing arrest rates are related to changes in police policies and practices or to changes in juvenile behavior. Comparing victim reports and arrest data to juvenile self-reports of behavior improves the situation somewhat. Many self-report studies, however, are conducted with school-based samples, omitting dropouts and truants who may have higher offending rates than children and adolescents who attend school regularly.
Although the panel acknowledges the weaknesses in available data, we nevertheless had to rely on currently available data to analyze juvenile crime trends. Based on our analysis, the panel drew the following conclusions.
There was an increase in juvenile homicide beginning in the mid-1980s, peaking in the early 1990s, and decreasing in the late 1990s. This increase was not confined to juveniles, however. For example, homicide