Sentencing Law and Policy

"What Explains Persistent Racial Disproportionality in Minnesota’s Prison and Jail Populations?"

The question in the headline of this post is the title of this important new article by Professor Richard Frase. Here is the abstract:

Racial disparity in prison and jail populations, measured by the ratio of black to white per capita incarceration rates, varies substantially from state to state.  To understand these variations, researchers must examine disparity at earlier stages of the criminal process and also racial differences in socioeconomic status that help explain disparity in cases entering the system. Researchers must adjust disparity ratios to correct for limitations in available data and in studies of prior incarceration rates.  Minnesota has one of the highest black/white incarceration ratios. Disparities at the earliest measurable stages of Minnesota’s criminal process — arrest and felony conviction — are as great as the disparity in total custody (prison plus jail) populations.  Disparities are substantially greater in prison sentences imposed and prison populations than at arrest and conviction.  The primary reason is the heavy weight sentencing guidelines give to offenders’ prior conviction records.  Highly disparate arrest rates appear to reflect unusually high rates of socioeconomic disparity between black and white residents.

"[A]mong those facing drug-related charges, incarceration and supervision seem not to deter subsequent criminal behavior"

The title of this post is the key take-away from the abstract of this new criminology paper on SSRN from Donald Green and Daniel Winik. The paper is titled "Using Random Judge Assignments to Estimate the Effects of Incarceration and Probation on Recidivism Among Drug Offenders," and here is the abstract:

Most prior studies of recidivism have used observational data to estimate the causal effect of imprisonment or probation on the probability that a convicted individual is re-arrested after release. Very few studies have taken advantage of the fact that in some jurisdictions, defendants are randomly assigned to judges who vary in sentencing tendencies.  The present study investigates whether defendants who are randomly assigned to more punitive judges have different recidivism probabilities than defendants who are assigned to relatively lenient judges.

We track 1,003 defendants charged with drug-related offenses (and no non-drug-related offenses) who were randomly assigned to nine judicial calendars between June 1, 2002 and May 9, 2003. Judges on these calendars meted out sentences that varied substantially in terms of prison and probation time. We tracked defendants using court records over a four-year period following the disposition of their cases in order to determine whether they were subsequently re-arrested.  Our results indicate that randomly-assigned variations in prison and probation time have no detectable effect on rates of re-arrest.  The findings suggest that, at least among those facing drug-related charges, incarceration and supervision seem not to deter subsequent criminal behavior.

If you build good guidelines, judges will follow

The title of this post seems to be the message of this local article discussing data on how frequently state judges in Pennsylvania sentence within applicable state sentencing guidelines.  Here are some of the details from an interesting (and all-too-rare) press discussion of state sentencing patterns:

There were 101,576 criminal sentences handed down in Pennsylvania courtrooms in 2008, and 1,530 of them were imposed by judges in the four Valley courts, according to the Pennsylvania Commission on Sentencing’s annual report.

Statewide, about 90 percent of the sentences meted out to convicted offenders conformed with the guidelines and 5 percent were shorter than suggested by the guidelines. In the Valley, the rate ranged from 92 percent conformity in Northumberland County to 98 percent in Montour County.

Judge William H. Wiest, one of three judges in Northumberland County, said he follows the state’s recommendations almost all the time because “they are well thought out, fair and reasonable.” Wiest said he’d like to see the guidelines broadened to allow judges the option of meting out harsher sentences.

After 11 years on the bench, he’d like to see the state increase its recommended sentencing for first-time drug sale offenders from a maximum of three months in jail to a maximum of seven months and allow more leeway in sentencing convicted drunken drivers. “I’d like a greater range of discretion,” he said....

Martin Wilson, assistant district attorney in Union County, said the creation of statewide sentencing guidelines helped level the playing field and ensured offenders in Philadelphia would face the same penalties as an offender who commits the same crime in the rural city of Sunbury. “They definitely served a valuable service in an effort to be consistent,” he said.

Notable empirical examination of truth-in-sentencing laws in operation

I just noticed on SSRN this article by two European economists titled "Responses to More Severe Punishment in the Courtroom: Evidence from Truth-in-Sentencing Laws." Here is the abstract:

We investigate behavioral responses of agents in the criminal process to more severe punishments by analyzing the effects of Truth-in-Sentencing (TIS) laws in a large sample of individual criminal cases.  The TIS laws raised the effective punishment by requiring offenders to serve at least 85% of their sentence in prison. Differences between the U.S. states in the timing of adoption and the types of crimes covered provide a source of identification.  The TIS laws increased the likelihood that the defendant would be acquitted or the case would be dismissed. The prosecutors responded by offering less advantageous concessions in plea bargaining -- they became less likely to reduce the charges and the defendants in turn became less likely to plead guilty.  The expected sentence for an arrested offender declined by more than 10% through a combination of changes in the probability of conviction and the sentence imposed upon conviction.  The potential deterrent effect of the TIS laws was therefore partially mitigated by the behavioral responses.

A timely examination of data and integration amidst California's corrections craziness

Professor W. David Ball has recently put up on SSRN on this timely analysisof California's sentencing and corrections challenges, which is titled "E Pluribus Unum: Data and Operations Integration in the California Criminal Justice System."  Here is the abstract:

The Stanford Criminal Justice Center (SCJC) recently completed a series of Executive Sessions with state and local officials about integrated criminal justice in California, exploring the ways in which the hundreds of disparate criminal justice agencies across the state might share information and coordinate activity, cooperating across jurisdictional and agency lines to promote common public safety goals.  An integrated criminal justice system, one where information is readily available to agencies when they need it, has several potential advantages: it can promote more efficient use of resources by avoiding duplication of effort; provide greater transparency to policymakers, regulatory agencies, and the public; and produce the evidence necessary to illustrate ways in which existing policies can be improved.

While integration is a crucial part of the future of criminal justice, integration itself is an increasingly important issue in its own right, particularly as governments tackle complex problems that do not confine themselves to particular geographic or jurisdictional areas (e.g. environmental pollution).  As with criminal justice, tackling these problems also requires massive amounts of information and inter-agency and inter-jurisdictional coordination.  Some lessons from the integrated criminal justice context might be relevant here: the importance of agreeing on common metrics, the challenge of getting individual agencies to think about how their information and interventions might be reused, and the importance of ensuring that any proposed changes take ordinary business practices into account. Integrated criminal justice can, at a minimum, illustrate the issues that are likely to arise.

Some recent posts on related issues in California and elsewhere:

• Federal judicial panel orders California to drastically cut prison population
• California begins SCOTUS appeal process for federal ruling ordering prisoner release
• Updates on all the prison craziness in California
• Editorial suggests why federal judges have had to be involved in California prison reform
• California's court-ordered plan for prison population reductions to come up short
• Details on California's not-quite-complete prison population reduction plan
• Economic necessity finally forcing long-needed reform in California
• "The Fiscal Crisis in Corrections: Rethinking Policies and Practices"
• Reviewing how tough times are resulting in prison releases
• "Prison spending still shackles state budget"
• Dueling persectives on proposed California prison reforms and a new challenge
• Trickle-down realities of the prison economy in California
• Latest news on sentencing and prison reform in California

Data bleg on gender, crime, victims and punishment

For various reasons, I have been thinking a lot lately about the gender realities in crime and punishment, and I am starting a new scholarly project that call for citing some historical and modern data relating to gender, crime, victims and punishment.  Thus, I thought it useful and wise to make this data bleg:

Can folks cite/link for me — either in the comments below or via e-mail — the best-known and/or most accurate and reliable historical and modern data breaking down with particularity gender dynamics surrounding criminal offending and crime victims? 

I would also be interesting in gendered data concerning sentencing and punishment for various crimes in various eras.  Thanks!

Interesting empirical study of sentencing in Nevada

I just came across this paper on SSRN, titled "Felony Sentencing in Rural and Urban Courts: Comparing Formal Legal and Substantive Political Models in the West." Here is the abstract:

This study explored two models of sentencing in urban and rural districts in Nevada (2007 felony sentencing data, N=10,873).  It was hypothesized that sentence lengths and dispositions would differ between rural and urban districts.  Furthermore, it was hypothesized that urban districts would follow a formal legal model – in which only legal variables (crime features, criminal history) act as predictors of sentencing outcomes, whereas rural districts would follow a substantive political model – in which extralegal factors (age, sex, race / ethnicity, etc.) also predict sentencing outcomes.  The authors found that urban and rural districts in Nevada conformed to a substantive political model of sentencing (including both legal and extralegal factors), though legal factors were often the strongest predictors of sentencing outcomes.

I have long thought that a lot more could be learned from the careful study of modern state sentencing systems, and this helpful paper confirms my desire to see more researchers analyzing more state sentencing patterns.

Good news on crime victimization rates in 2008

Various folks expressed concern that the down-turn in the economy in 2008 could and likely would lead to an increase in crimes.  But, as this DOJ press release details, the latest, greatest statistics suggest that crime victimization decreased from 2007 to 2008.  Here is the encouraging news from the start of the press release:

The violent crime rate in 2008 — 19.3 victimizations per 1,000 persons age 12 or older — was unchanged from the previous year, the Bureau of Justice Statistics (BJS) in the Office of Justice Programs, U.S. Department of Justice, announced today.  The property crime rate declined during 2008 from 147 to 135 crimes per 1,000 households, primarily as a result of decreases in theft and motor vehicle theft.

In 2008, an estimated 4.9 million violent crimes (rapes or sexual assaults, robberies, aggravated assaults and simple assaults) occurred, as well as an estimated 16.3 million property crimes (burglaries, motor vehicle thefts and household thefts) and 137,000 personal thefts (picked pockets and snatched purses).  These offenses included both crimes reported and unreported to police.  With the exception of theft and motor vehicle theft, victimization rates for every type of crime measured were unchanged from the 2007 levels.

Violent and property crime rates in 2008 remain at the lowest levels recorded since 1973, the first year that such data were collected.  The rate of every major violent and property crime measured by BJS fell between 1999 and 2008.  The overall violent crime rate fell 41 percent and the property crime rate declined by 32 percent during the last 10 years.

The full statistical report is available in this new publication, which is titled "Criminal Victimization, 2008."  Now we can begin the usual debate over who and what accounts for (1) the fact that crime is now at record low levels in the US, and (2) the fact that there are still, on average, tens of thousands of serious crimes every single day in the US.

Letter to the USSC about modern sentencing data

I had the honor of signing my name to a terrific letter put together last week by a group of sentencing gurus that urged the US Sentencing Commission to go further in making its sentencing data more transparent and accessible. This letter can be downloaded below, and here is how it begins:

The undersigned scholars and researchers study the federal sentencing system from a variety of policy and philosophical perspectives.  As you well know, these are interesting times. Constitutional decisions by the Supreme Court revised the guideline system put in place by the Sentencing Reform Act of 1984 (the SRA).  Legislation, both enacted and proposed, has raised important questions concerning the fairness and effectiveness of federal sentencing.  The quality of our national policy debate can be improved by empirical research, just as our sentencing practices can be improved by empirical evaluation.

We call on you to fully use the powers granted by the SRA to facilitate research and evaluation. The upcoming Data and Research Conference on May 5-7 provides an opportunity for the Commission to expand its role as “clearinghouse and information center” on federal sentencing practices and to assist researchers working with the annual monitoring data file.  The conference agenda published on your website does not seem to address several concerns regarding the timing and method of release of Commission data, however, and the release of data beyond the annual monitoring, appeals, and organizations data files.  We ask that you consider these additional issues and recommendations.

Download Open_letter_to_USSC_April_20

Would "spreading the wealth" help reduce crime and punishment rates?

This new little criminology paper that just showed up on SSRN, titled "Income Inequality and Crime in the United States," ought to help generate a robust debate on a friday.  Here is the brief abstract of the brief paper:

This paper investigates the relationship between income inequality and crime.  Results show that there is a strong and robust effect of relative income inequality on burglary.  Effect on robbery is also strong and robust in most cases.

Of course, these results linking economic crime to economic conditions should hardly be surprising.  But, in light of the apparent reality that all sorts of crime and punishment levels often also to correlate to inequalities in wealth (and educational attainment), one cannot help but wonder if Senator Jim Webb's new proposal to "create a blue-ribbon commission to look at every aspect of our criminal justice system with an eye toward reshaping the process from top to bottom" will have to both start and end with "spread the wealth" recommendations.

US Sentencing Commission to conduct a data and research conference in May

I just noticed on the US Sentencing Commission's website this very exciting news: "The United States Sentencing Commission will present a data and research conference in Washington, D.C. on May 5-7, 2009."  This link provides these additional details:

This conference will help researchers use Commission data by demonstrating how to find and access Commission’s data and research as well as providing instructional help in working with the Commission’s individual offender datafiles.  The conference will be held in the Federal Judicial Center Auditorium of the Thurgood Marshall Federal Judiciary Building in Washington, D.C. near Union Station and Capitol Hill.

The conference will provide attendees with an indepth understanding of how the Commission’s individual offender data is assembled, including how the data is collected from documents submitted to the Commission and how the data is checked for accuracy.  Commission staff will also provide training on how to use specific fields within the datafile and will discuss special problems in using federal sentencing data.  The conference will also highlight recent research by Commission staff using Commission data.  The conference schedule is available at http://www.ussc.gov/DataConf/Schedule.htm.

This is fantastic news about a much-needed (and long-overdue) effort by the US Sentencing Commission to help researchers of all stripes take advantage of all the amazing data and information that the USSC has collected and coded over the last two decades.  And the official schedule suggests that both data junkies and data neophytes like me should be able to learn a lot at this important and timely event.

A new opportunity to assess the economy/crime rate connection

The relationship between economic conditions and crime rates has always been considered an important (and unclear) component of criminal justice realities.  This interesting new article from today's Denver Post, which is headlined "This time, economy, crime fall in tandem: Metro area seems to buck trend of downturns breeding scofflaws," provides some new food for thoughts concerning this relationship.  Here is how the article starts:

The number of reported burglaries, car thefts and larcenies has dropped steadily in several Colorado cities despite historic data that say crime should be on the rise during this deep economic recession.

The decrease puzzles criminologists but gives police reason to tout crime-prevention initiatives and tougher sentencing laws putting prolific burglars and car thieves away. Yet even the police aren't sure of all of the factors driving the phenomenon. "Perhaps it would be a natural assumption that crime would increase when the economy is bad," said Detective Shannon Lucy of the Aurora Police Department. "It does kind of go against what people would expect."

It may be too early in this recession to see whether the downward crime trend holds. Some who are young, poor and uneducated may yet be driven to crime after repeated failed attempts to find jobs, some experts said. "People don't become criminals overnight. It takes some time for the strain to hit them," said Jeff London, assistant professor of criminology at Metropolitan State College of Denver.

Between 1979 and 1981, when wages dropped 20 percent, the number of property crimes reported jumped 18 percent nationwide, following a long-established pattern in which recessions cause crime to spike. Colorado experienced a similar leap of 18 percent in property-crime numbers between 1979 and 1981, according to Colorado Department of Public Safety reports.

But during the current recession, several Colorado communities, including Denver, Thornton and Aurora, are seeing double-digit-percentage declines. Statewide totals for 2008 won't be available for two months.

But while the numbers look promising for getting through the recession without a crime spike, there are dominoes yet to fall. Mass layoffs during a recession force people with master's degrees to seek jobs normally taken by those with bachelor's degrees, who, in turn, take openings normally taken by less-educated youths, who find it increasingly difficult to find any job. It also takes time for people to run out of unemployment benefits.

US Sentencing Commission releases 2008 Annual Report and Sourcebook

Though this weekend I likely will have a severe case of March Madness with lots of thinking about RPI data, the US Sentencing Commission has just released a set of materials on its website that would also enable me to have a severe case of Guidelines madness with lots of thinking about sentencing data.  Specifically, here's what now available for download from the USSC's website:

2008 Annual Report and Sourcebook: The 2008 Annual Report presents an overview of major Commission activities and accomplishments for fiscal year 2008.  See the Commission's 2008 Sourcebook of Federal Sentencing Statistics for descriptive figures, tables, and charts, and selected district, circuit, and national sentencing data.

Latest, greatest federal sentencing data run from USSC

The US Sentencing Commission has some new post-Kimbrough/Gall sentencing data now up on its website. The USSC's updated data report, which can be accessed here, is described this way:

Fiscal Year 2008 Post-Kimbrough/Gall Data Report: An updated set of tables presenting final data on fiscal year 2008 cases sentenced on or after December 10, 2007 through September 30, 2008. This report was prepared using data received, coded, and edited by the Commission by February 10, 2009.

Notably, the new data continue to show that ust under 60% of all post-Gall sentences are now within the calculated guidelines range, though this fact is mostly a product of prosecutors requesting a below-range sentence in more than 25% of all cases.  And, as has always been the reality in the federal sentencing system, one can identify a number of large inter-circuit and inter-district variations in how many sentences fall within or outside calculated guideline ranges.

Two great new data reports from the US Sentencing Commission

I am pleased to discover and report that the US Sentencing Commission has used the holiday season to produce two lovely new (and very reader-friendly) documents about federal sentencing realities.  First came last week's release of "Overview of Federal Criminal Cases, Fiscal Year 2007," which is actually a more exciting document than the title might suggest.  Here is how the USSC describes this report:

This publication provides a broad overview of federal sentencing data for fiscal year 2007. Readers will find this publication to be a brief, easy-to-use reference on the types of criminal cases handled by the federal courts and the punishments imposed on the offenders convicted in those cases. This publication will be available in hard copy format in the coming weeks.

And this week has now brought another report, "Changing Face of Federal Criminal Sentencing: Seventeen Years of Growth in the Federal Sentencing Caseload."  Though perhaps only a true sentencing geek like me finds even this title exciting, here is how the USSC whets our appetite for this notable new document:

This publication provides an overview of the demographic make-up of federal offenders sentenced under the federal sentencing guidelines for fiscal years 1991 through 2007, examines trends in the types of federal offenses sentenced, and analyzes the demographic characteristics of offenders sentenced for the most common federal offenses during this time period.

I now know at least one data nerd who can make special plans for a rocking federal sentencing new year's eve bash.  (Maybe I should drop all my USSC data runs from my roof at the stroke of midnight to celebrate all the great information the USSC has published in 2008.  It truly has been a banner year for federal sentencing data junkies.)

Lots of holiday gifts from BJS for data junkies

This is the time of year in which the Bureau of Justice Statistics releases a whole bunch of statistics about prisoners and others under state and federal criminal control.  This year I got an e-mail from BJS providing links to a cornucopia of data.  Overwhelmed by all the goodies, I decided I'll just cut-and-paste all the goodies and links from the e-mail:

Prisoners in 2007
Available online at: http://www.ojp.usdoj.gov/bjs/abstract/p07.htm.

Presents data on prisoners under jurisdiction of federal or state
correctional authorities on December 31, 2007, collected from the
National Prisoner Statistics series.


Probation and Parole in the United States, 2007 - Statistical Tables
Available online at: http://www.ojp.usdoj.gov/bjs/abstract/ppus07st.htm.

Presents the number of persons on probation and parole at yearend 2007,
by state, with percent changes in each state during the year.


BJS Key Facts at a Glance charts

Incarceration rates, 1980-2007
Available at: http://www.ojp.usdoj.gov/bjs/glance/incrt.htm

State prison population by offense type, 1980-2005
Available at: http://www.ojp.usdoj.gov/bjs/glance/corrtyp.htm

Adult correctional populations, 1980-2007
Available at: http://www.ojp.usdoj.gov/bjs/glance/corr2.htm

Latest BJS census of correctional facilities

A new report just released by the Bureau of Justice Statistics, titled "Census of State and Federal Correctional Facilities, 2005" provides an accounting of nationwide facilities locking up individuals in the Unites States.  The new BJS report is available online at this link, where these highlights are noted:

• The number of prisoners held in custody in state and federal correctional facilities increased 10% from 1,305,253 in 2000 to 1,430,208 in 2005.
• Between the 2000 and the 2005 censuses, the number of correctional employees rose 3%, resulting in a higher inmate-to-staff ratio in the latter year.
• While the stock of minimum security facilities grew by 155 and maximum security facilities rose by 40 between 2000 and 2005, the number of medium security facilities declined by 42.

New BJS data on federal criminal justice realities

A new report just released by the Bureau of Justice Statistics provides a different perspective on federal crime and punishment than is provided by the data produced by the US Sentencing Commission. The new BJS report, titled simply "Federal Justice Statistics, 2005" and available online at this link, provides arrest data and other case-processing information that provides a more complete view of the massive realities of the federal criminal justice system.

This official press release provides an overview of some of the many interesting particulars appearing in the full BJS report:

Immigration and drug arrests comprised more than half of the 140,200 federal suspects arrested and booked by the U.S. Marshals in 2005, the Justice Department’s Bureau of Justice Statistics (BJS) announced today....

Forty percent of all suspects arrested by the U.S. Marshals Service were arrested in 1 of 5 federal judicial districts along the U.S.-Mexico border, including Arizona, New Mexico, the Southern and Western Districts of Texas, and the Southern District of California. Nearly 1 in 4 (23 percent) of all suspects arrested in 2005 were arrested in the Southern and Western Districts of Texas....

From 1995 to 2005, the likelihood of being prosecuted, convicted, and sentenced to prison increased. Sixty percent of suspects referred to U.S. attorneys were prosecuted in 2005 (from 54 percent in 1995) Ninety percent of defendants charged with a federal violation were convicted (up from 84 percent in 1995) and 79 percent of defendants convicted in 2005 were sentenced to prison (up from 67 percent in 1995). Thirteen percent of the defendants convicted in 2005 were sentenced to probation (down from 24 percent in 1995)....

The U.S. Court of Appeals received 14,644 filings in 2005 — an increase of 44 percent from 1995.  About 4 in 10 appeals filed in 2005 were for drug offenses (41 percent), followed by immigration (17 percent), weapons (16 percent) and violent (5 percent) offenses.  Appeals filed by immigration offenders increased an average 25 percent per year from 1995 to 2005.

     At year-end 2005, 375,600 persons were under some form of federal supervision — 62 percent were in secure confinement, including prison and pretrial detention. Thirty-eight percent were under federal supervision in the community, including pretrial release and post-sentencing supervision.