The research examined three files. The prosecutor's data were collected from the Polk County prosecutor's office in Des Moines, Iowa. Data were based on a 10-percent systematic sample of cases closed between July 1, 1979, and June 30, 1980 (334 cases). Public defender data were obtained from offices in Kentucky and Nebraska, both of which use the National Legal Aid and Defenders Association's AMICUS system for recording attorney effort times. For all three files, time is recorded for an attorney activity as it occurs. Data analysis identified major factors determining attorney effort and productivity in a given case: the dispositional route of the case, the type of offense, the severity of the sanction attached to the case, and the defendant's criminal history. Thus, levels of attorney effort are variable and potentially subject to change, but different levels of productivity (number of cases processed for a given amount of attorney hours) can be achieved by distributing the workload in various ways. Changes in productivity were examined under a 10-percent reduction in trials in Kentucky and Nebraska along with a 10-percent increase in crime in Iowa. Effects of these changes on productivity are calculated. The author notes that the unanswered question is to what extent resources can be redistributed for increased productivity without violating the quality of justice dispensed. Future research is suggested for developing definitions of capacity, efficiency, quality of output, and equity. Twenty-three footnotes as well as tabular and graphic data are provided.