Legislative Waiver Reconsidered: General Deterrent Effects of Statutory Exclusion Laws Enacted Post‐1979

Abstract

In recent years, the juvenile justice system has undergone a “get tough” transformation. One component of the system which has experienced substantial alteration is the mechanism by which states allow for the waiver of juvenile offenders to adult criminal court. Most of the state systems now have some form of transfer procedure in place and many allow for the automatic transfer (or statutory exclusion) of juveniles who have been charged with certain offenses. Although the effects of waiver laws on individuals have received much empirical attention, their effects on the respective states’ aggregate level violent juvenile crime rates are less understood. In this study, we examine the relative effects of legislative waiver laws in 22 states that have added statutory exclusion provisions since 1979. In doing so, we assess whether legislative waiver should be reconsidered.

Keywords:

• waiver
• transfer
• juvenile
• deterrence

Acknowledgments

An earlier draft of this paper was presented at the Annual Conference of the American Society of Criminology, November 16–20, 2004, Nashville, TN. The data sets examined for this paper were made available by the International Consortium for Political and Social Research. The data for “Uniform Crime Reports: Monthly Weapon Specific Crime and Arrest Time Series 1975–1993” (ICPSR 6792) were collected by Susan Carlson. The data for the “Uniform Crime Reporting Series” (ICSPR 2004‐02‐24, 2000‐06‐21, 2001‐06‐29, 2002‐06‐27, 2003‐06‐05, and 2004‐06‐04) were collected by the Federal Bureau of Investigation. Neither the collectors of the original data nor the Consortium bear any responsibility for the analyses presented here. The authors would like to thank Mitchell Chamlin for his helpful comments during the analysis of the data.

Notes

1. “Substantially amended” was defined as expanding their law from including solely murder to incorporating additional offenses. We did not examine eight states that already had legislative waiver laws in place in 1979. Two other states (Florida and New Hampshire) were excluded from our analysis because of data collection issues. Florida added a legislative waiver statute in 2000; however, data were unavailable for all of the months surrounding the statutes’ effective date. New Hampshire added a legislative waiver provision in 2002. Accordingly, there were not enough observations after the law went into effect to assess its effects.

2. Kentucky’s legislative waiver law went into effect January 1, 2001. As a result, data were only available for 2 years after the law went into effect. Attempts were made to attain more recent data from the state of Kentucky; however, the data were unavailable.

3. We contacted Dr. Carlson to assess the validity of the state‐level data. Dr. Carlson advised that univariate ARIMA predictions models were used to estimate some of the state‐level data for two states (Michigan and Texas), neither of which were assessed here.

4. From 1991 to the present, Alabama did not report monthly arrest data to the FBI. Accordingly, the data for Alabama were obtained directly from the Alabama State Criminal Justice Information Center. In 1991 Iowa did not report monthly arrest data to the FBI. The Iowa data were obtained from the Iowa Department of Public Safety. Missouri’s legislative waiver law went into effect January 1, 1999. We obtained Missouri’s 2003 data, which was not yet available from the FBI, directly from the Missouri State Highway Patrol.

5. A significant first order transfer function was estimated for Maine. However, after examining the δ parameter it was determined that the zero order transfer function was more appropriate (see McDowell et al., 1980).

6. A significant pulse function was estimated for Missouri. However, after examining the δ parameter it was determined that the first order transfer function was more appropriate (see McDowall et al., 1980).

7. Because 2003 monthly data were not yet available the control series were modeled with only 4 years of observations after the intervention. A Nebraska pulse model approached significance (t = 1.51, p < .13) and might be worth reassessing when the data become available.

8. These parameters and final models are available from the corresponding author upon request.

Additional information

Notes on contributors

Benjamin Steiner

Benjamin Steiner is a Research Associate and doctoral candidate in the Division of Criminal Justice at the University of Cincinnati. He holds an MA from Boise State University.

Craig Hemmens

Craig Hemmens is Chair and Professor in the Department of Criminal Justice Administration and Academic Director of the Paralegal Studies Program at Boise State University. He holds a JD from North Carolina Central University School of Law and a PhD in Criminal Justice from Sam Houston State University. He has published eight books and more than 100 articles on a variety of topics related to criminal justice.

Valerie Bell

Valerie Bell is an instructor at Boise State University. She has an MA in Criminal Justice Administration from Boise State University.