Gender, Status, & Tax Offenses Over Time

ABSTRACT

Men are more likely than women to commit tax offenses, and previous research has suggested that this difference can be partially explained by status processes. This study investigates whether the relationship between status and tax offenses has changed over time in synchrony with changes in social stratification based on gender, race, and class. We hypothesize that if gender equality has increased within U.S. society (particularly within the formal economy), then gender-based discrepancies in tax offending should have decreased as well. Using administrative data from federal criminal cases sentenced between 1995–2015, we examine how the effects of different indicators of status – including race, class, and gender – on the odds of being convicted for a tax offense have changed over time. Results from a series of logistic regression models indicate that over time the difference in the probability of men and women being convicted of a tax offense versus another white collar crime decreases. Other variables of status including race, age, and education are also analyzed and discussed.

Acknowledgments

The authors wish to thank Dr. Holly Foster for her support and valuable comments on all prior and current work which led to the development of this paper.

Notes

¹ While we would have liked to examine each type of tax offense separately, the types of offenses were only disaggregated between 1991–1998, and during those years, the majority of offenses fell within one of three different categories: tax evasion, failure to pay, or false statements. During all other years, tax offenses were aggregated under a single category. As a result, we compare the characteristics of all tax offenders to all other white collar criminals.

² Although Reese and McDougal (2017) included income within their analysis, data pertaining to defendants’ annual incomes was only available through 1994–1995 and not available for analysis during the remaining time frame of this study.

³ Although we report results from this analysis, we also validated these results by running the same analysis using listwise deletion to handle cases of missing data. Results from these analyses were comparable to those produced when using multiple imputations.

⁴ When we used listwise deletion to account for cases with missing data, the incorporation of additional measures of status decreased the size of the effect of gender on status in 19 out of 20 years. In 2011, the size of the effect of gender remained the same but the direction of the effect reversed after incorporating additional measures of status. The odds of being convicted for a tax offense compared to any other white collar crime were 8% higher for women (rather than men) in 2011.

Additional information

Notes on contributors

Bruce Reese

Bruce Reese is an assistant professor and the director of the Criminal Justice/Criminology department at Concordia University. He received his PhD in Sociology from Texas A&M University, his M.A. in Psychology from Boston University, and his B.A. in Psychology from Tufts University. His broad interests are in social psychology, criminology, deviance, and inequality. His research primarily focuses on cooperation, sanctioning, and prosocial behavior.

Katie Constantin

Katie Constantin is a graduate student in the Department of Sociology at Texas A&M University. She received her B.A. in Anthropology/Sociology from Rhodes College and her M.A. in Sociology from the University of New Orleans. Her primary areas of interest include social psychology, small groups, punishment, and violence/victimization. Her research focuses on cooperation and punishment in public goods games as well as sexual violence and victimization.