The relation between gender and criminality is strong, and is likely to remain so. Women have traditionally been much less likely than men to commit violent crimes, and that pattern persists today. Rates of female involvement in some forms of property crime-notably petty theft and fraud-appear to be increasing. However, while the relative increase in women's property crime involvement is significant, female participation even in these crimes remains far less than that of men. The relation of gender to case processing decisions in the criminal justice system varies from stage to stage. Although the pertinent literature is plagued by methodological and interpretive problems, several tentative conclusions can be offered. Women are more likely than men, other things equal, to be released on recognizance; however, when bail is set, the amount of bail does not appear to be affected by the defendant's gender. There is no clear evidence that the defendant's gender systematically affects prosecution, plea negotiation, or conviction decisions. In sentencing, however, women appear to receive systematic leniency except when they are convicted of high-severity offenses.

This study analyzes data for state criminal defendants prosecuted in New York to determine the bases upon which judges make pretrial release decisions for these defendants. Treating statutory law as defining the category of legal variables, it finds legal factors substantially affect decisions about whether to release a defendant on recognizance, the amount of bail required, and whether to offer a defendant a cash alternative to a surety bond. The impact of these factors varies, however, depending upon the particular decision being made. Factors not prescribed in the statute-extra-legal factors-are also found to affect these pretrial release decisions. Their impact, too, is decision context specific. Among the extra-legal factors that affect pretrial release decisions, the effects of status characteristics of the defendant pale in comparison to the effects of bench bias and measures of the defendant's dangerousness.

female offenders have been at the margins of the criminal justice system. Theories of criminal behavior, as well as studies of arrest, pre-trial, prosecution, and sentencing outcomes, have tended to focus on patterns of criminality derived from studying male offenders.' This does not reflect a lack of interest in female offenders, but rather the empirical fact that the vast majority of criminal offenders, especially violent criminal offenders, have been male. 2 In other words, the traditional preoccupation of theorists, researchers, and criminal justice professionals with male offenders derives from the gender-skewed demographics of criminal behavior. Recently, however, the combination of the women's rights movement, the rise of feminist scholarship, and the noted increase in female criminality, 3 has begun to reverse this long-standing neglect of * Member of the United States Sentencing Commission and Professor of Law at Indiana University (Bloomington) School of Law. The views expressed here are those of the individual authors and are not meant to represent the views of the United States Sentencing Commission.

As early as 1939, the perception of white-collar crime as a national problem had emerged. Edwin Sutherland, for example, compared the crime and punishment of the white-collar class, composed of respected businessmen and professionals, with that of the lower class. 1 His work, which he viewed less as a call for policy reform than as a contribution to theoretical criminology, stimulated succeeding decades of research. A consistent theme, whether stated explicitly or implicitly, runs through this body of research: laws and sanctions are differentially applied, to the benefit of the white-collar offender. 2 Sutherland argued that this differential treatment did not reflect important distinctions between the types of offenses committed by the two cla:sses.

No abstract