Wednesday, August 09, 2017

Still more interesting new "Quick Facts" publications on federal drug sentencing from the US Sentencing Commission

In this post a few month ago, I noted that the US Sentencing Commission had released a notable new Quick Facts covering all "Drug Trafficking Offenses"  (As the USSC explains and reglar readers know, "Quick Facts" are official publications that "give readers basic facts about a single area of federal crime in an easy-to-read, two-page format.")   Now I see that the USSC has just released this big set of new Quick Facts covering individual drugs:

The data appearing these publications runs through Fiscal Year 2016, which is end of September 2016, and thus they set something of a benchmark for the end of the Obama era before the start of the Trump era of federal criminal policies and practices.

August 9, 2017 in Data on sentencing, Detailed sentencing data, Drug Offense Sentencing, Federal Sentencing Guidelines | Permalink | Comments (0)

Sunday, July 30, 2017

Should an uptick in federal gun prosecutions garner bipartisan praise?

The question in the title of this post was my first thought upon seeing this press release from the Justice Department released Friday under the heading "Federal Gun Prosecutions Up 23 Percent After Sessions Memo."  Here is the full text of the press release:

Today, the U.S. Department of Justice announced that, following the memorandum from Attorney General Sessions to prioritize firearm prosecutions, the number of defendants charged with unlawful possession of a firearm increased nearly 23 percent in the second quarter of 2017 (2,637) from the same time period in 2016 (2,149).

“Violent crime is on the rise in many parts of this country, with 27 of our biggest 35 cities in the country coping with rising homicide rates,” said Attorney General Jeff Sessions.  “Law abiding people in some of these communities are living in fear, as they see families torn apart and young lives cut short by gangs and drug traffickers.  Following President Trump’s Executive Order to focus on reducing crime, I directed federal prosecutors to prioritize taking illegal guns off of our streets, and as a result, we are now prosecuting hundreds more firearms defendants. In the first three months since the memo went into effect, charges of unlawful possession of a gun -- mostly by previously convicted felons -- are up by 23 percent.  That sends a clear message to criminals all over this country that if you carry a gun illegally, you will be held accountable.  I am grateful to the many federal prosecutors and agents who are working hard every day to make America safe again.”

In February, immediately after the swearing-in of Attorney General Jeff Sessions, President Trump signed an Executive Order that directs the Attorney General to seek to reduce crime and to set up the Task Force on Crime Reduction and Public Safety.  The Task Force has provided Sessions with recommendations on a rolling basis.  In March, based on these recommendations, Attorney General Sessions sent a memorandum to Department of Justice prosecutors, ordering them to prioritize firearms offenses.

In the three months immediately following the Attorney General’s memo -- April, May and June -- the number of defendants charged with unlawful possession of a firearm (18 U.S.C. 922) increased by nearly 23 percent compared to those charged over the same time period in 2016.  The number of defendants charged with the crime of using a firearm in a crime of violence or drug trafficking (18 U.S.C. 924), increased by 10 percent.

Based on data from the Executive Office for United States Attorneys (EOUSA), in Fiscal Year 2016 (starting October 1), 11,656 defendants were charged with firearms offenses under 18 U.S.C. 922 or 924.  EOUSA projects that in Fiscal Year 2017, the Department is on pace to charge 12,626 defendants with these firearms crimes.  That would be the most federal firearms cases since 2005.  It would also be an increase of eight percent from Fiscal Year 2016, 20 percent from 2015, and an increase of 23 percent from 2014.

Of course, as regular readers on this blog know well, many on the political left have been critical of various efforts by AG Sessions to ramp up federal prosecutions. But much of the criticism is based on concerns about escalating the federal drug war, especially as it applies to lower-lever and nonviolent offenders. As the title of this post is meant to suggest, perhaps this latest data showing a ramp up of gun prosecutions could be met with some applause from political left given the tendency of the left to support tougher restrictions on gun possession. (Of course, some parts of the libertarian-faction of the political right has also expressed concerns about recent work by AG Sessions, and they might be more troubled by these data.)

Critically, without having more information about the "who and how" of increased federal gun prosecutions, I do not feel sufficiently informed to robustly praise or criticize these developments. But I do think it interesting and notably that the first new data being stressed by the Sessions DOJ involves a type of prosecution that could garner support from both sides of the political aisle.

July 30, 2017 in Data on sentencing, Detailed sentencing data, Federal Sentencing Guidelines, Gun policy and sentencing, Who Sentences? | Permalink | Comments (9)

Thursday, July 27, 2017

Helpful summary of USSC's recent overview of mandatory minimums in federal system

As reported in this prior post, the US Sentencing Commission earlier this month released a lengthy new publication titled "An Overview of Mandatory Minimum Penalties in the Federal Criminal Justice System" reviewing the use of federal mandatory minimum penalties and their impact on the federal prison population. Now the USSC has released this two-page publication with the big report's highlights. Here are highlights of these highlights:

July 27, 2017 in Data on sentencing, Detailed sentencing data, Mandatory minimum sentencing statutes | Permalink | Comments (1)

Thursday, July 20, 2017

"The Immediate Consequences of Pretrial Detention: Evidence from Federal Criminal Cases"

The title of this post is the title of this interesting empirical paper authored by Stephanie Holmes Didwania that was recently posted on SSRN. Here is the abstract:

This paper presents evidence of the effects of pretrial detention status on criminal case outcomes in federal criminal cases. I find that criminal defendants who are released pending trial earn a roughly 72 percent decrease in sentence length and a 36 percentage-point increase in the probability of receiving a sentence below the recommended federal sentencing Guidelines range. Pretrial release also reduces the probability that a defendant will receive at least the mandatory minimum sentence — when one is charged — by 39 percentage points, but does not affect the probability that the defendant will face a mandatory minimum sentence.

To address the identification problem inherent in using pretrial detention status as an explanatory variable, I take advantage of the fact that pretrial release in federal courts is typically determined by magistrate judges who vary in their propensities to release defendants pending trial. This setting allows magistrate judge leniency to serve as an instrumental variable for pretrial release. I also present suggestive evidence of the mechanism at work. It appears that pretrial release affects case outcomes in two distinct ways: most importantly, by giving defendants the opportunity to present mitigating evidence at sentencing and, secondly, by making it easier for defendants to earn a sentencing reduction by providing substantial assistance to the government. In contrast, this paper does not find evidence that pretrial release improves defendants’ abilities to bargain with prosecutors. I also find that the effects of pretrial detention status on case outcomes are heterogeneous, and most pronounced for drug offenders.

July 20, 2017 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Procedure and Proof at Sentencing | Permalink | Comments (1)

Thursday, July 13, 2017

Urban Institute releases "A Matter of Time: The Causes and Consequences of Rising Time Served in America’s Prisons"

Logo-simpleThis morning the Urban Institute released online here a big new project on long prison terms titled, "A Matter of Time: The Causes and Consequences of Rising Time Served in America’s Prisons." As explained in an email I received, this "online feature examines the causes and consequences of rising time served in America’s prisons [t]hrough visualizations, analysis of trends and demographics, and stories told by people who have served long prison terms." An executive summary can be found at this link, and here are excerpts from it:

People are spending more time in prison, and the longest prison terms are getting longer.  Since 2000, average time served has risen in all 44 states that reported complete data to the National Corrections Reporting Program.  In states with more extensive data, we can trace the rise back to the 1980s and 1990s. In nearly half the states we looked at, the average length of the top 10 percent of prison terms increased by more than five years between 2000 and 2014.

The increase in time served has been sharpest among people convicted of violent offenses.  These changes have an outsized effect on prison populations because people convicted of violent offenses make up more than half the people in state prisons and the majority of people with long terms.

Longer terms are growing in number and as a share of the prison population.  In 35 states, at least 1 in 10 people in prison have been there for a decade or more.  This is even higher — nearly 1 in 4 people — in states like California and Michigan.  In at least 11 states, the number of people who have served at least a decade has more than doubled since 2000.

These trends aren’t accidental, and that they vary so much across states suggests that the growth in time served is driven by state-level decisionmaking.  States grappling with expanding prison populations must include those serving the longest prison terms in their efforts to curb mass incarceration.

Incarceration affects some people and communities more than others, and these patterns are often more pronounced among those who spend the most time in prison.  In 35 of the 44 states we looked at, racial disparities in prisons were starkest among people serving the longest 10 percent of terms.  In recent years, racial disparities have decreased among people serving less than 10 years, but 18 states actually saw an increase in disparities among people serving longer terms.

Nearly two in five people serving the longest prison terms were incarcerated before age 25, despite research that shows the brain is still developing through age 24 and that people tend to age out of criminal behavior.  Thousands have been in prison for more than half their lives.  One in five people in prison for at least 10 years is a black man incarcerated before age 25.

A growing share of women in prison have served more than 10 years.  In Michigan, for example, 8 percent of women in prison had served at least a decade as of 2000; by 2013, that number was 13 percent.  In Wisconsin, this figure rose from 1.8 to 6.5 percent over the same period.  In light of this trend, more research is needed to understand how women are uniquely affected by long-term incarceration.

More than one in three people serving the longest prison terms is at least 55 years old.  More people serving longer terms means that more people are growing old in prison, yet prisons are typically ill-equipped to address the needs of the elderly and disabled.

July 13, 2017 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Scope of Imprisonment | Permalink | Comments (1)

Tuesday, July 11, 2017

US Sentencing Commission releases new overview of mandatory minimums in federal system

As reported in this official press release, the "United States Sentencing Commission today released a new publication — An Overview of Mandatory Minimum Penalties in the Federal Criminal Justice System (2017 Overview) — that examines the use of federal mandatory minimum penalties and the impact of those penalties on the federal prison population." Here is more from the press release about this new publication and its findings:

The new publication updates much of the data contained in its 2011 Report to the Congress: Mandatory Minimum Penalties in the Federal Criminal Justice System and compiles data through 2016, the most recent full fiscal year for which federal sentencing data is available.

Judge William H. Pryor, Jr., Acting Chair of the Commission stated, "This publication examines the latest data about the use of mandatory minimum sentences in the federal criminal justice system.  When Congress created the Commission, Congress empowered it to serve 'as a clearinghouse and information center' about federal sentencing and to assist Congress, the federal courts, and federal departments in the development of sound sentencing policies.  See 28 U.S.C. § 995(a)(12)(A). The Commission has published this report to fulfill that Congressional mandate."

Among the key data findings in the publication are:

  • The average sentence length for federal offenders convicted of an offense carrying a mandatory minimum penalty in fiscal year 2016 was 110 months of prison, nearly four times the average sentence (28 months) for offenders whose offense did not carry a mandatory minimum.

  • Slightly more than half (55.7%) of federal inmates in custody as of September 30, 2016 were convicted of an offense carrying a mandatory minimum.

  • Over one-third (38.7%) of federal offenders convicted of an offense carrying a mandatory minimum penalty in fiscal year 2016 received relief from the mandatory minimum at sentencing, which is a decrease from 46.7 percent in fiscal year 2010.

  • Hispanic offenders continued to represent the largest group of federal offenders (40.4%) convicted of an offense carrying a mandatory minimum penalty in fiscal year 2016.

  • White offenders had the longest average sentence (127 months) among federal offenders convicted of an offense carrying a mandatory minimum penalty in fiscal year 2016, which is a shift from fiscal year 2010 when Black offenders convicted of an offense carrying a mandatory minimum penalty had the longest average sentence (127 months).

  • While Black offenders convicted of an offense carrying a mandatory minimum penalty continued to receive relief from the mandatory minimum penalty least often, the gap between Black offenders and White offenders has narrowed from a difference of 11.6 percent in fiscal year 2010 to 3.2 percent in fiscal year 2016.

The 2017 Overview is part of a multi-year study included in the Commission’s policy priorities over the past several amendment cycles and is intended to be the first in a series of reports on mandatory minimum penalties.  Continuation of the study is listed as a tentative policy priority for the amendment year ending May 1, 2018.  The Commission will accept public comment on proposed priorities through July 31, 2017.

The full USSC report, which runs 89 pages, is available at this link. I hope to find some time in the coming weeks to highlights some additional data from this latest review of the latest mandatory minimum realities.

July 11, 2017 in Data on sentencing, Detailed sentencing data, Mandatory minimum sentencing statutes | Permalink | Comments (2)

Thursday, June 08, 2017

More interesting new "Quick Facts" publications from the US Sentencing Commission

The US Sentencing Commission has released two notable new Quick Facts covering "Drug Trafficking Offenses" and "Federal Offenders in Prison" as of February 2017. (As the USSC explains, "Quick Facts" are publications that "give readers basic facts about a single area of federal crime in an easy-to-read, two-page format.")  Here are a few of the many intriguing data details from these two small data-filled publications:

June 8, 2017 in Data on sentencing, Detailed sentencing data, Drug Offense Sentencing, Federal Sentencing Guidelines, Mandatory minimum sentencing statutes, Prisons and prisoners, Race, Class, and Gender, Scope of Imprisonment | Permalink | Comments (2)

Wednesday, May 31, 2017

"Era of Mass Expansion: Why State Officials Should Fight Jail Growth"

Era_MassIncarceration_250The title of this post is the title of this notable new report from the Prison Policy Initiative. This PPI press release provide context and summary for the report's coverage:

State capitols share responsibility for growing jail populations, charges a new report by the Prison Policy Initiative. “Jails are ostensibly locally controlled, but the people held there are generally accused of violating state law, and all too often, state policymakers ignore jails,” argues the new report, Era of Mass Expansion: Why State Officials Should Fight Jail Growth.

The fact that jails are smaller than state prison systems and under local control has allowed state officials to avoid addressing the problems arising from jail policies and practices. “Reducing the number of people jailed has obvious benefits for individuals, but also helps states curb prison growth down the line,” says Joshua Aiken, report author and Policy Fellow at the Prison Policy Initiative.

Every year, 11 million people churn through local jail systems, mostly for minor violations of state law. Of the 720,000 people in jails on a given day, most have not been convicted of a crime and have either just been arrested or are too poor to make bail. And since the 1980s, crime has fallen but the number of people jailed has more than tripled.

The new report finds that the key driver of jail growth is not what one might expect – courtroom judges finding more people guilty and sentencing them to jail. In fact, the number of people serving jail sentences has actually fallen over the last 20 years. Instead, the report finds two troubling explanations for jail growth:

  • An increasing number of people held pre-trial.

  • Growing demand from federal and state agencies to rent cell space from local jails.

Recognizing the importance of state-specific data for policymakers and advocates, the report offers more than a hundred graphs that make possible state comparisons of jail trends. The report uncovers unique state problems that drive mass incarceration:

  • In some states, state officials have not utilized their ability to regulate the commercial bail industry, which has profited from the increased reliance on money bail and increased bail amounts. These trends have expanded the pre-trial population dramatically over time.

  • In other states, state lawmakers have expanded criminal codes, enabled overzealous prosecutors, and allowed police practices to play a paramount role in driving up jail populations, while underfunding pre-trial programs and alternatives to incarceration.

  • In 25 states, 10% or more of the people confined in local jails are being held for state or federal agencies, with some counties even adding capacity to meet the demand. This report is the first to be able to address the local jail population separately from the troubling issue of renting jail space.

Era of Mass Expansion draws particular attention to the states where the dubious practice of renting jail space to other authorities contributes most to jail growth. “Local sheriffs, especially in states like Louisiana and Kentucky, end up running a side business of incarcerating people for the state prison system or immigration authorities,” explains Aiken. “Renting out jail space often creates a financial incentive to expand jail facilities and keep more people behind bars.” The report finds that renting jail space for profit has contributed more to national jail growth since the 1980s than people who are being held by local authorities and who have actually been convicted of crimes.

For state policymakers, the report offers 10 specific recommendations to change how offenses are classified and treated by law enforcement, eliminate policies that criminalize poverty or create financial incentives for unnecessarily punitive practices, and monitor the upstream effects of local discretion. “There are plenty of things local officials can do to lower the jail population,” says Aiken. “With this report, I wanted to bring in state-level actors by showing how much of the solution is in their hands.”

May 31, 2017 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Scope of Imprisonment, Who Sentences? | Permalink | Comments (5)

Friday, May 26, 2017

US Sentencing Commission releases report on "Youthful Offenders in the Federal System"

Cover_youthful-offendersThe US Sentencing Commission released this notable new report today titled simply "Youthful Offenders in the Federal System." Here is the report's introduction and "key findings" from its first two pages:

Introduction

Although youthful offenders account for about 18 percent of all federal offenders sentenced between fiscal years 2010 and 2015, there is little current information published about them.  In this publication, the United States Sentencing Commission (“the Commission”) presents information about youthful offenders, who for purposes of this report are defined as persons age 25 or younger at the time they are sentenced in the federal system.

Recent studies on brain development and age, coupled with recent Supreme Court decisions recognizing differences in offender culpability due to age, have led some policymakers to reconsider how youthful offenders should be punished.  This report reviews those studies and provides an overview of youthful federal offenders, including their demographic characteristics, what type of offenses they were sentenced for, how they were sentenced, and the extent of their criminal histories.

The report also discusses the intersection of neuroscience and law, and how this intersection has influenced the treatment of youthful offenders in the criminal justice system. The Commission is releasing this report as part of its review of the sentencing of youthful offenders.  In June 2016, the Commission’s Tribal Issues Advisory Group (TIAG) issued a report that proposed several guideline and policy changes relating to youthful offenders, including departure provisions and alternatives to incarceration.

Because many of the TIAG recommendations on this topic apply to all youthful offenders, and not just Native Americans, the Commission voted to study the treatment of youthful offenders as a policy priority for the 2016-2017 amendment cycle.

The key findings in this report are that:

• There were 86,309 offenders (18.0% of the federal offender population) age 25 or younger sentenced in the federal system between 2010 and 2015.

• The majority (57.8%) of youthful offenders are Hispanic.

• There were very few youthful offenders under the age of 18 sentenced in the federal system (52 between 2010 and 2015).

• Almost 92 percent of offenses committed by youthful offenders were nonviolent offenses.

• Similar to the overall federal offender population (or non-youthful offenders group) the most common offenses that youthful offenders committed were drug trafficking (30.9%), immigration (28.6%), and firearms offenses (13.7%).

• The average sentence for youthful offenders was 34.9 months.

• Youthful offenders were more likely to be sentenced within the guidelines range than non-youthful offenders (56.1% compared to 50.1%).

• Youthful offenders recidivated at a much higher rate than their older counterparts — about 67 percent versus 41 percent.

May 26, 2017 in Data on sentencing, Detailed sentencing data, Offender Characteristics, Race, Class, and Gender | Permalink | Comments (3)

"U.S. Prison Population Trends 1999-2015: Modest Reductions with Significant Variation"

The title of this post is the title of this brief "Fact Sheet" from The Sentencing Project, which gets started this way:

While states and the federal government have modestly reduced their prison populations in recent years, incarceration trends continue to vary significantly across jurisdictions. Overall, the number of people held in state and federal prisons has declined by 4.9% since reaching its peak in 2009.  Sixteen states have achieved double-digit rates of decline and the federal system has downsized at almost twice the national rate.  But while 38 states have reduced their prison populations, in most states this change has been relatively modest.  In addition, 12 states have continued to expand their prison populations even though most have shared in the nationwide crime drop.

Six states have reduced their prison populations by over 20% since reaching their peak levels:

• New Jersey (35% decline since 1999)

• New York (29% decline since 1999)

• Alaska (27% decline since 2006)

• California (26% decline since 2006, though partly offset by increasing jail use)

• Vermont (25% decline since 2009)

• Connecticut (22% decline since 2007)

Southern states including Mississippi, South Carolina, and Louisiana, which have exceptionally high rates of incarceration, have also begun to significantly downsize their prison populations.  These reductions have come about through a mix of changes in policy and practice designed to reduce admissions to prison and lengths of stay.  Moreover, the states with the most substantial prison population reductions have often outpaced the nationwide crime drop.

The pace of decarceration has been very modest in most states, especially given that nationwide violent and property crime rates have fallen by half since 1991.  Despite often sharing in these crime trends, 15 states had less than a 5% prison population decline since their peak year.  Moreover, 12 states have continued to expand their prison populations, with four producing doubledigit increases since 2010: North Dakota, Wyoming, Oklahoma, and Minnesota.

May 26, 2017 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Scope of Imprisonment, State Sentencing Guidelines | Permalink | Comments (0)

Thursday, May 25, 2017

"The portal itself is like a video game for criminal justice nerds."

The title of this post is a sentence that got me (too) excited about this new criminal justice resource called "Measure for Justice."  The sentence is found in this Marshall Project article, fully headlined "The New Tool That Could Revolutionize How We Measure Justice: A small nonprofit gathers criminal justice statistics, one county at a time." Here are excerpts from the article describing what is revolutionary about Measures for Justice:

The enormity of the country’s criminal justice system — 15,000 state and local courts, 18,000 local law enforcement agencies, more than two million prisoners — looks even more daunting when you consider how little we know about what is actually going on in there. Want to know who we prosecute and why? Good luck. Curious about how many people are charged with misdemeanors each year? Can’t tell you. How about how many people reoffend after prison? We don’t really know that, either.

In an age when everything is measured — when data determines the television we watch, the clothes we buy and the posts we see on Facebook — the justice system is a disturbing exception. Agencies exist in silos, and their data stays with them. Instead, we make policy based on anecdote, heavily filtered through a political lens.

This week the nonprofit Measures for Justice is launching an online tool meant to shine a high beam into these dark corners. It is gathering numbers from key criminal justice players — prosecutors offices, public defenders, courts, probation departments — in each of America’s more than 3,000 counties. Staffers clean the data, assemble it in an apples-to-apples format, use it to answer a standard set of basic questions, and make the results free and easy to access and understand.

It’s the kind of task you’d expect a federal agency like the Bureau of Justice Statistics, which has an average annual budget of $97 million, to take on. Instead, the 22 people at Measures for Justice’s Rochester, N.Y., offices are doing the work themselves on an annual budget of $4.6 million, donated mostly from foundations. So far they’ve tackled six states: Washington, Utah, Wisconsin, Pennsylvania, North Carolina and Florida, gathering most of the numbers one county at a time. Together, these make up 10 percent of the nation’s counties. The team chose those six states for their geographical diversity and — to ease the data gathering in the project’s early phases — because they had unified statewide court databases. The hope is to complete 15 more states by 2020, while updating the statistics from the first six states every two years.

“We’re giving people data they’ve never had access to before,” says Amy Bach, the founder and executive director of Measures for Justice. “We’re telling them stories about their communities and their counties that they’ve never heard before.” The project, which has as its motto “you can’t change what you can’t see,” centers on 32 “core measures”: yardsticks to determine how well local criminal justice systems are working. How often do people plead guilty without a lawyer? How often do prosecutors dismiss charges? How long do people have to wait for a court hearing? Users can also slice the answers to these questions in different ways, using “companion measures” such as race and political affiliation.

The portal itself is like a video game for criminal justice nerds. Users can compare counties, click on interactive maps and bar charts, and layer one data point upon another. The interface is clean and easy to use, if a little wonky. (The organization wants to present data in context, so each infographic is followed by a screen full of fine print and footnotes.) It’s meant for everyone — not just professors and policy wonks.

May 25, 2017 in Data on sentencing, Detailed sentencing data, Who Sentences? | Permalink | Comments (0)

Wednesday, May 24, 2017

"The Price of Prisons: Examining State Spending Trends, 2010 - 2015"

Images (2)The title of this post is the title of this notable new Vera Institute of Justice report. Here is its introduction:

After decades of a stable rate of incarceration, the U.S. prison population experienced unprecedented growth from the early 1970s into the new millennium — with the number of people confined to state prisons increasing by more than 600 percent, reaching just over 1.4 million people by the end of 2009.The engine driving this growth was the enactment and implementation over time of a broad array of tough-on-crime policies, including the rapid and continuous expansion of the criminal code; the adoption of zero-tolerance policing tactics, particularly around minor street-level drug and quality-of-life offenses; and the proliferation of harsh sentencing and release policies aimed at keeping people in prison for longer periods of time (such as mandatory minimum sentences, truth-in-sentencing statutes, and habitual offender laws).

Unsurprisingly, this explosion in the use of incarceration had a direct financial influence on state budgets.  Creating and sustaining such a sprawling penal system has been expensive.  With more people under their care, state prison systems were compelled to build new prison facilities and expand existing ones.  To staff these new and expanded facilities, they also had to hire, train, and retain ever more employees.  In addition to expanding the state-operated prison system, some states also began to board out increasing numbers of people to county jails, privately-run facilities, and other states’ prison systems.

After hitting a high of 1.4 million people in 2009, however, the overall state prison population has since declined by 5 percent, or 77,000 people.  Lawmakers in nearly every state and from across the political spectrum — some prompted by the 2008 recession — have enacted new laws to reduce prison populations and spending, often guided by a now-large body of research supporting alternative, more effective responses to crime.7 In addition to fiscal pressures, the push for reform has been further bolstered by other factors, including low crime rates; shifting public opinion that now favors less incarceration and more rehabilitation; and dissatisfaction with past punitive policies that have failed to moderate persistently high recidivism rates among those sent to prison.

With these various political, institutional, and economic forces at play, most states have adopted a variety of different policies, including those that increase opportunities to divert people away from the traditional criminal justice process; expand the use of community-based sanctions; reduce the length and severity of prison sentences for certain offenses, including the rollback of mandatory penalties; increase opportunities for people to gain early release; and better provide enhanced reentry support for those leaving prison or jail.

In light of nearly a decade of broad-based criminal justice reform, this report seeks to determine where state prison spending stands today and how it has changed in recent years.  In particular, if a goal of recent reforms has been to make deep and lasting cuts to prison spending by reducing the prison population, have states who have witnessed the desired downward shift in prison size also witnessed it in spending?  To answer this question, researchers at the Vera Institute of Justice (Vera) developed a survey to measure changes in state prison population and expenditures between 2010 and 2015, and conducted follow-up interviews with state prison budget officials to better understand spending and population trends.

Vera’s study confirms that prisons remain an expensive enterprise, despite the success of many states — including Michigan, New Jersey, New York, and South Carolina —in simultaneously reducing their prison populations while achieving budget savings.  The first part of this report describes 2015 prison expenditures, identifying the main driver of corrections spending across responding states.  The second half of the report then discusses how changes in prison populations during the study period, and other trends largely outside the control of departments of corrections have affected prison spending. What is clear is that increased spending is not inevitable, since nearly half of states have cut their spending on prisons between 2010 and 2015.  But while one might expect that states with shrinking prison populations are uniformly spending less on prisons, or conversely that states with growing populations are spending more, Vera’s findings paint a more complicated picture.  Indeed, often there is no single reason that explains a rise or fall in spending, but a multitude of factors that push and pull expenditures in different directions.

May 24, 2017 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Scope of Imprisonment | Permalink | Comments (3)

Monday, May 08, 2017

US Sentencing Commission releases report providing overview of FY 2016 federal sentencing cases

Fig1_fy16overviewThe US Sentencing Commission just released this helpful and relatively brief data report titled simply "Overview of Federal Criminal Cases Fiscal Year 2016."  Among other useful realities, this report provides a certain kind of data marker for the end of the "Obama era" for federal caseload and sentencing patterns.  (The chart reprinted here from the report shows how the number of persons federal sentenced significantly increased during Obama's first term and significantly decreased during Obama's second term.) Here is the overview of the USSC report and key findings via this USSC webpage:

The United States Sentencing Commission received information on 67,874 federal criminal cases in which the offender was sentenced in fiscal year 2016. Among these cases, 67,742 involved an individual offender and 132 involved a corporation or other “organizational” offender. The Commission also received information on 11,991 cases in which the court resentenced the offender or modified the sentence that had been previously imposed. This publication provides an overview of those cases.

Key Findings

A review of cases reported to the Commission in fiscal year 2016 reveal the following:

  • The 67,742 individual original cases reported to the Commission in fiscal year 2016 represent a decrease of 21.4% since fiscal year 2011, the year in which the largest number of offenders were sentenced. Drug cases continued to be the most common type of federal case, accounting for 31.6% of all cases.

  • Methamphetamine offenses continued to be the most common drug cases, representing 30.8% of all drug crimes.  The proportion of methamphetamine cases has increased substantially since fiscal year 1994, when those cases accounted for only 6.4% of all drug cases.

  • Just under half (44.5%) of all drug offenders were convicted of an offense carrying a mandatory minimum penalty; however, this proportion was the lowest it has been since fiscal year 1993.

  • Immigration cases were the second most common, accounting for 29.6% of the total federal caseload.  In fiscal year 2011, immigration cases were the most common federal crime — however, since that year the number of these cases has steadily declined.

  • Crimes involving firearms were the third most common offense, accounting for 10.8% of the total number of federal criminal convictions in fiscal year 2016.  The average sentence imposed in firearms cases was 75 months.

  • There were 6,517 fraud cases in fiscal year 2016, accounting for 9.6% of the total federal caseload; however, this number represents a 12.2% reduction from the year before.

May 8, 2017 in Booker in district courts, Data on sentencing, Detailed sentencing data, Federal Sentencing Guidelines | Permalink | Comments (0)

Thursday, May 04, 2017

"The Use of Risk Assessment at Sentencing: Implications for Research and Policy"

The title of this post is the title of this paper authored by Jordan Hyatt and Steven Chanenson recently posted to SSRN. Here is the abstract:

At-sentencing risk assessments are predictions of an individual’s statistically likely future criminal conduct.  These assessments can be derived from a number of methodologies ranging from unstructured clinical judgment to advanced statistical and actuarial processes.  Some assessments consider only correlates of criminal recidivism, while others also take into account criminogenic needs.  Assessments of this nature have long been used to classify defendants for treatment and supervision within prisons and on community supervision, but they have only relatively recently begun to be used — or considered for use — during the sentencing process.  This shift in application has raised substantial practical and policy challenges and questions.

This paper, supported by the U.S. Department of Justice’s Bureau of Justice Assistance, directly addresses these issues and provides information and examples from a range of jurisdictions, including some which have integrated at-sentencing risk assessment programs in place or are in the process of doing so.  Derived from a survey of judges, as well as a series of interviews with stakeholders from across the nation, opportunities for future research and planning to guide the cautious engagement with at-sentencing risk assessment are identified.

May 4, 2017 in Data on sentencing, Detailed sentencing data, Procedure and Proof at Sentencing, Purposes of Punishment and Sentencing, Who Sentences? | Permalink | Comments (0)

The Sentencing Project reports on "America’s Increasing Use of Life and Long-Term Sentences"

The Sentencing Project yesterday released this significant new report titled "Still Life: America’s Increasing Use of Life and Long-Term Sentences." Here is its introduction:

The number of people serving life sentences in U.S. prisons is at an all-time high. Nearly 162,000 people are serving a life sentence -- one of every nine people in prison.  An additional 44,311 individuals are serving “virtual life” sentences of 50 years or more. Incorporating this category of life sentence, the total population serving a life or virtual life sentence reached 206,268 in 2016.  This represents 13.9 percent of the prison population, or one of every seven people behind bars. A mix of factors has led to the broad use of life sentences in the United States, placing it in stark contrast to practices in other nations.

Every state and the federal government allow prison sentences that are so long that death in prison is presumed.  This report provides a comprehensive profile of those living in this deep end of the justice system. Our analysis provides current figures on people serving life with parole (LWP) and life without parole (LWOP) as well as a category of long-term prisoner that has not previously been quantified: those serving “virtual” or de facto life sentences.  Even though virtual life sentences can extend beyond the typical lifespan, because the sentences are not legally considered life sentences, traditional counts of life-sentenced prisoners have excluded them until now.

KEY FINDINGS

• As of 2016, there were 161,957 people serving life sentences, or one of every nine people in prison.

• An additional 44,311 individuals are serving “virtual life” sentences, yielding a total population of life and virtual life sentences at 206,268 – or one of every seven people in prison.

• The pool of people serving life sentences has more than quadrupled since 1984.The increase in the LWOP population has far outpaced the changes in the LWP population.

• There are 44,311 people serving prison sentences that are 50 years or longer. In Indiana, Louisiana, and Montana, more than 11 percent of the prison population is serving a de facto life sentence.

• Nearly half (48.3%) of life and virtual life-sentenced individuals are African American, equal to one in five black prisoners overall.

• Nearly 12,000 people have been sentenced to life or virtual life for crimes committed as juveniles; of these over 2,300 were sentenced to life without parole.

• More than 17,000 individuals with an LWP, LWOP, or virtual life sentence have been convicted of nonviolent crimes.

• The United States incarcerates people for life at a rate of 50 per 100,000, roughly equivalent to the entire incarceration rates of the Scandinavian nations of Denmark, Finland, and Sweden.

May 4, 2017 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Scope of Imprisonment | Permalink | Comments (6)

Monday, May 01, 2017

Spotlighting again the use of risk-assessment computations at sentencing (under an inaccurate headline)

Adam Liptak has this new column discussing the Loomis risk-assessment sentencing case pending SCOTUS cert review, but the column bears the inaccurate headline "Sent to Prison by a Software Program’s Secret Algorithms."  As of this writing, software programs alone have not sent any persons to prison, not in the Wisconsin case before SCOTUS or any other that I know about.  Software may be making recommendations to sentencing decision-makers, and that certainly justifies scrutiny, but we have not quite yet reached the brave new world that this headline suggests.  That said, the headline did grab my attention, and here are parts of the article that follows:

[A] Wisconsin man, Eric L. Loomis, who was sentenced to six years in prison based in part on a private company’s proprietary software. Mr. Loomis says his right to due process was violated by a judge’s consideration of a report generated by the software’s secret algorithm, one Mr. Loomis was unable to inspect or challenge.

In March, in a signal that the justices were intrigued by Mr. Loomis’s case, they asked the federal government to file a friend-of-the-court brief offering its views on whether the court should hear his appeal.

The report in Mr. Loomis’s case was produced by a product called Compas, sold by Northpointe Inc. It included a series of bar charts that assessed the risk that Mr. Loomis would commit more crimes. The Compas report, a prosecutor told the trial judge, showed “a high risk of violence, high risk of recidivism, high pretrial risk.” The judge agreed, telling Mr. Loomis that “you’re identified, through the Compas assessment, as an individual who is a high risk to the community.”

The Wisconsin Supreme Court ruled against Mr. Loomis. The report added valuable information, it said, and Mr. Loomis would have gotten the same sentence based solely on the usual factors, including his crime — fleeing the police in a car — and his criminal history.

At the same time, the court seemed uneasy with using a secret algorithm to send a man to prison. Justice Ann Walsh Bradley, writing for the court, discussed, for instance, a report from ProPublica about Compas that concluded that black defendants in Broward County, Fla., “were far more likely than white defendants to be incorrectly judged to be at a higher rate of recidivism.”

Justice Bradley noted that Northpointe had disputed the analysis. Still, she wrote, “this study and others raise concerns regarding how a Compas assessment’s risk factors correlate with race.” In the end, though, Justice Bradley allowed sentencing judges to use Compas. They must take account of the algorithm’s limitations and the secrecy surrounding it, she wrote, but said the software could be helpful “in providing the sentencing court with as much information as possible in order to arrive at an individualized sentence.”

Justice Bradley made Compas’s role in sentencing sound like the consideration of race in a selective university’s holistic admissions program. It could be one factor among many, she wrote, but not the determinative one.

In urging the United States Supreme Court not to hear the case, Wisconsin’s attorney general, Brad D. Schimel, seemed to acknowledge that the questions in the case were substantial ones. But he said the justices should not move too fast. “The use of risk assessments by sentencing courts is a novel issue, which needs time for further percolation,” Mr. Schimel wrote.

He added that Mr. Loomis “was free to question the assessment and explain its possible flaws.” But it is a little hard to see how he could do that without access to the algorithm itself. The company that markets Compas says its formula is a trade secret. “The key to our product is the algorithms, and they’re proprietary,” one of its executives said last year. “We’ve created them, and we don’t release them because it’s certainly a core piece of our business.”

Compas and other products with similar algorithms play a role in many states’ criminal justice systems. “These proprietary techniques are used to set bail, determine sentences, and even contribute to determinations about guilt or innocence,” a report from the Electronic Privacy Information Center found [available here]. “Yet the inner workings of these tools are largely hidden from public view.”...

There are good reasons to use data to ensure uniformity in sentencing. It is less clear that uniformity must come at the price of secrecy, particularly when the justification for secrecy is the protection of a private company’s profits. The government can surely develop its own algorithms and allow defense lawyers to evaluate them. At Rensselaer last month, Chief Justice Roberts said that judges had work to do in an era of rapid change. “The impact of technology has been across the board,” he said, “and we haven’t yet really absorbed how it’s going to change the way we do business.” 

Some prior related posts on Loomis case:

May 1, 2017 in Data on sentencing, Detailed sentencing data, Procedure and Proof at Sentencing, Purposes of Punishment and Sentencing, Who Sentences? | Permalink | Comments (3)

Thursday, April 27, 2017

"Mandatory Minimum Policy Reform and the Sentencing of Crack Cocaine Defendants: An Analysis of the Fair Sentencing Act"

The title of this post is the title of this notable new article by David Bjerk just published by the Journal of Empirical Studies. Here is the abstract:

The Fair Sentencing Act of 2010 (FSA) affected the U.S. federal mandatory minimum sentencing laws for crack cocaine offenders, and represented the first congressional reform of sentencing laws in over 20 years.  A primary goal of this legislation was to lessen the harshness of sentences for crack cocaine offenders and decrease the sentencing gap between crack defendants and powder cocaine defendants.  While the mean sentence length for crack offenders fell following the implementation of the FSA, these changes appear to primarily reflect the continuation of ongoing sentencing trends that were initiated by a variety of noncongressional reforms to federal sentencing policy that commenced around 2007.  However, the FSA appears to have been helpful in allowing these trends to continue past 2010.

April 27, 2017 in Booker in district courts, Data on sentencing, Detailed sentencing data, Drug Offense Sentencing, Federal Sentencing Guidelines, Mandatory minimum sentencing statutes, New crack statute and the FSA's impact | Permalink | Comments (0)

Wednesday, March 29, 2017

Notable review of recent ups and downs in federal prosecutions

FT_17.03.23_prosecutions_numberIn this new posting over at the Pew Research Center, John Gramlich has assembled interesting data on federal modern criminal justice realities under the headline "Federal criminal prosecutions fall to lowest level in nearly two decades." Here are highlights:

After peaking in 2011, the number of federal criminal prosecutions has declined for five consecutive years and is now at its lowest level in nearly two decades, according to a Pew Research Center analysis of new data from the federal court system. The decline comes as Attorney General Jeff Sessions has indicated that the Justice Department will reverse the trend and ramp up criminal prosecutions in the years ahead.

Federal prosecutors filed criminal charges against 77,152 defendants in fiscal year 2016, according to the Administrative Office of the U.S. Courts. That’s a decline of 25% since fiscal 2011, when 102,617 defendants were charged, and marks the lowest yearly total since 1997. The data count all defendants charged in U.S. district courts with felonies and serious misdemeanors, as well as some defendants charged with petty offenses. They exclude defendants whose cases were handled by magistrate judges.

Prosecutions for drug, immigration and property offenses – the three most common categories of crime charged by the federal government – all have declined over the past five years. The Justice Department filed drug charges against 24,638 defendants in 2016, down 23% from 2011. It filed immigration charges against 20,762 defendants, down 26%. And it charged 10,712 people with property offenses such as fraud and embezzlement, a 39% decline.

However, prosecutions for other, less frequently charged crime types have increased slightly. For example, prosecutors charged 8,576 defendants with gun crimes in 2016, a 3% increase over 2011 (and a 9% single-year increase over 2015). And they charged 2,897 people with violent crimes such as murder, robbery and assault, a 4% increase from five years earlier.

Several factors may play a role in the decline in federal prosecutions in recent years. One notable shift came in 2013, when then-Attorney General Eric Holder directed federal prosecutors to ensure that each case they bring “serves a substantial federal interest.” In a speech announcing the policy change, Holder said prosecutors “cannot – and should not – bring every case or charge every defendant who stands accused of violating federal law.”

Sessions, who took office as attorney general in February, has indicated that the Justice Department will take a different approach under his leadership. In particular, he has pushed to increase prosecutions for drug- and gun-related offenses as part of a broader plan to reduce violent crime, which rose nationally in 2015 and in the first half of 2016, according to the FBI. (Despite these increases, violent crime remains far below the levels recorded in the 1990s.)...

Since 2001, the Justice Department’s prosecution priorities have changed. Immigration offenses, for instance, comprised just 15% of all prosecutions in 2001; by 2016, they accounted for 27%. During the same period, drug crimes fell from 38% to 32% of all prosecutions, while property crimes declined from 20% to 14%.

Such revisions by the Justice Department are not unusual. In 2013, for example, after two states legalized the recreational use of marijuana, the department announced new charging priorities for offenses involving the drug, which remains illegal under federal law. Federal marijuana prosecutions fell to 5,158 in 2016, down 39% from five years earlier.

March 29, 2017 in Data on sentencing, Detailed sentencing data, Federal Sentencing Guidelines, Offense Characteristics, Who Sentences? | Permalink | Comments (2)

Tuesday, March 14, 2017

Prison Policy Initiative releases 2017 version of "Mass Incarceration: The Whole Pie"

Pie2017The Prison Policy Initiative has an updated version of its terrific incarceration "pie" graphic and report, which is available at this link (along with a larger version of the pie graphic reprinted here). Here is part of the report's introductory text and subsequent discussion:

Wait, does the United States have 1.3 million or more than 2 million people in prison? Are most people in state and federal prisons locked up for drug offenses? Frustrating questions like these abound because our systems of confinement are so fragmented and controlled by various entities. There is a lot of interesting and valuable research out there, but varying definitions make it hard — for both people new to criminal justice and for experienced policy wonks — to get the big picture.

This report offers some much needed clarity by piecing together this country’s disparate systems of confinement. The American criminal justice system holds more than 2.3 million people in 1,719 state prisons, 102 federal prisons, 901 juvenile correctional facilities, 3,163 local jails, and 76 Indian Country jails as well as in military prisons, immigration detention facilities, civil commitment centers, and prisons in the U.S. territories. And we go deeper to provide further detail on why people are locked up in all of those different types of facilities.

Pie chart showing the number of people locked up on a given day in the United States by facility type and the underlying offense using the newest data available in March 2017.Pie chart showing the number of people locked up on a given day in the United States in jails, by convicted and not convicted status, and by the underlying offense, using the newest data available in March 2017. Graph showing the number of people in jails from 1983 to 2014 by whether they have been convicted or not. The number of convicted people stopped growing in 1999, but the number of unconvicted people continues to grow.Graph showing, for the years 2007 to 2015, the number of people ~~ 10.9 to 13.6 million ~~ a year who are admitted to jail per year and the number of people ~~ about 700,000 to 800,000 ~~ who are in jail on a given day.Graph showing the incarcerated populations in federal prisons, state prisons, and local jails from 1925 to 2015. The state prison and jail populations grew exponentially in the 1980s and 1990s, and began to decline slowly after 2008, while federal prison populations have always been smaller and show less change over time.

While this pie chart provides a comprehensive snapshot of our correctional system, the graphic does not capture the enormous churn in and out of our correctional facilities and the far larger universe of people whose lives are affected by the criminal justice system. Every year, 641,000 people walk out of prison gates, but people go to jail over 11 million times each year. Jail churn is particularly high because most people in jails have not been convicted. Some have just been arrested and will make bail in the next few hours or days, and others are too poor to make bail and must remain behind bars until their trial. Only a small number (187,000 on any given day) have been convicted, generally serving misdemeanors sentences under a year....

With a sense of the big picture, a common follow-up question might be: how many people are locked up for a drug offense? We know that almost half a million people are locked up because of a drug offense. The data confirms that nonviolent drug convictions are a defining characteristic of the federal prison system, but play only a supporting role at the state and local levels. While most people in state and local facilities are not locked up for drug offenses, most states’ continued practice of arresting people for drug possession destabilizes individual lives and communities. Drug arrests give residents of over-policed communities criminal records, which then reduce employment prospects and increase the likelihood of longer sentences for any future offenses.

All of the offense data presented comes with an important set of caveats. A person in prison for multiple offenses is reported only for the most serious offense so, for example, there are people in prison for “violent” offenses who might have also been convicted of a drug offense. Further, almost all convictions are the result of plea bargains, where people plead guilty to a lesser offense, perhaps of a different category or one that they may not have actually committed.

And many of these categories group together people convicted of a wide range of offenses. For example, “murder” is generally considered to be an extremely serious offense, but “murder” groups together the rare group of serial killers, with people who committed acts that are unlikely for reasons of circumstance or advanced age to ever happen again, with offenses that the average American may not consider to be murder at all. For example, the felony murder rule says that if someone dies during the commission of a felony, everyone involved can be as guilty of murder as the person who pulled the trigger. Driving a getaway car during a bank robbery where someone was accidentally killed is indeed a serious offense, but many may be surprised that this is considered murder.

March 14, 2017 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Scope of Imprisonment, Who Sentences? | Permalink | Comments (0)

Monday, March 13, 2017

US Sentencing Commission releases 2016 Annual Report and Sourcebook of Federal Sentencing Statistics

Via email I received this cursory report on the publication of lots of federal sentencing data that is anything but cursory:

The United States Sentencing Commission’s 2016 Annual Report and 2016 Sourcebook of Federal Sentencing Statistics are now available online.

The Annual Report provides an overview of the Commission’s activities and accomplishments in fiscal year 2016.

The Sourcebook of Federal Sentencing Statistics presents tables, figures, and charts on selected district, circuit, and national sentencing data for fiscal year 2016. The Commission collected and analyzed data from approximately 315,000 court documents for nearly 68,000 federal criminal cases in the production of this year’s Sourcebook.

I am hoping to find time to churn over a lot of the data in these reports, but already from the start of the 2016 Annual Report these data items jumped out:

March 13, 2017 in Data on sentencing, Detailed sentencing data, Federal Sentencing Guidelines | Permalink | Comments (2)

Friday, March 10, 2017

More interesting new Quick Facts on fraud sentencing from the US Sentencing Commission

I noted in this post earlier this week that the US Sentencing Commission had released the first of a new series of Quick Facts covering federal fraud sentencing with a focus on health care fraud cases. (As the USSC explains, "Quick Facts" are publications that "give readers basic facts about a single area of federal crime in an easy-to-read, two-page format.")  I have now just noticed that the USSC released a number of other fraud-focused Quick Facts this week, and here are links to them:

Hard-core federal sentencing fans might make a parlor game of trying to guess which type of fraud has the most and which has the least sentences imposed within the calculated guideline ranges.

March 10, 2017 in Data on sentencing, Detailed sentencing data, Federal Sentencing Guidelines, White-collar sentencing | Permalink | Comments (4)

Thursday, March 09, 2017

US Sentencing Commission releases another big recidivism report on federal offenders

The United States Sentencing Commission is continuing to publish important new data report about the recidivism rates and patterns of federal offenders.  This latest 44-page report is titled "The Past Predicts the Future: Criminal History and Recidivism of Federal Offenders."  This page on the USSC's website provides this summary and highlights:

The Past Predicts the Future: Criminal History and Recidivism of Federal Offenders examines a group of 25,431 federal offenders who were released from prison or placed on probation in calendar year 2005.  Information about the components of Chapter Four of the Guidelines Manual — including total criminal history score, criminal history category, and point assignments for types of past convictions — and their association with recidivism are contained in this report. The findings included in this report build on those in the Commission’s 2016 Recidivism Overview report.

Report Highlights

  • Consistent with its previous work in this area, the Commission found that recidivism rates are closely correlated with total criminal history points and resulting Criminal History Category classification, as offenders with lower criminal history scores have lower recidivism rates than offenders with higher criminal history scores.

  • The Commission found substantial differences in recidivism rates among Criminal History Category I offenders (which includes offenders with a criminal history score of zero or one point).  Less than one-third (30.2%) of Criminal History Category I offenders with zero points were rearrested while nearly half (46.9%) of offenders with one point were rearrested.

  • The Commission also found differences in recidivism rates among offenders with zero criminal history points. Offenders with zero points and no prior contact with the criminal justice system have a lower recidivism rate (25.7%) than offenders with zero points but some prior contact with the criminal justice system (37.4%).

  • Offenders who have less serious prior convictions (assigned one point) have a lower recidivism rate (53.4%) than offenders who have prior convictions assigned two or three points (71.3% for offenders with at least one two-point offense and 70.5% for offenders with at least one three-point offense).

March 9, 2017 in Data on sentencing, Detailed sentencing data, National and State Crime Data | Permalink | Comments (3)

Monday, March 06, 2017

Interesting new Quick Facts on federal health care fraud sentencing from the US Sentencing Commission

The US Sentencing Commission has released this notable new Quick Facts covering federal sentencing in health care fraud cases. (As the USSC explains, "Quick Facts" are publications that "give readers basic facts about a single area of federal crime in an easy-to-read, two-page format.")  Here are a few of the intriguing data details from the the publication highlighting that within-guideline sentencing is actually the exception rather than the norm in these cases:

During the past three years, the rate of within range sentences for health care fraud offenders has decreased from 43.6% in fiscal year 2013 to 32.9% in fiscal year 2015.

In each of the past three years, approximately one-fifth to one-third of health care fraud offenders received a sentence below the applicable guideline range because the government sponsored the below range sentence....

In each of the past three years, approximately 34 percent of health care fraud offenders received a non-government sponsored below range sentence.

March 6, 2017 in Data on sentencing, Detailed sentencing data, Federal Sentencing Guidelines, White-collar sentencing | Permalink | Comments (0)

Tuesday, February 21, 2017

US Sentencing Commission releases big new report on "Recidivism Among Federal Drug Trafficking Offenders"

The US Sentencing Commission today released the second major report emerging from a huge assessment of federal offenders released from prison in 2005.  This USSC webpage provides this background and highlights from this 149-page data-rich report:

This report, Recidivism Among Federal Drug Trafficking Offenders examines a group of 10,888 federal drug trafficking offenders who were released in calendar year 2005. These 10,888 offenders, who were all U.S. citizens, represent 42.8 percent of the 25,431 federal offenders who were released in calendar year 2005 and analyzed in the Commission’s 2016 report, Recidivism Among Federal Offenders: A Comprehensive Overview. In the future, the Commission will release additional publications discussing specific topics concerning recidivism of federal offenders.

Chapter One summarizes the group studied in this report as well as its key findings. It also explains the methodology used in the report. Chapter Two provides an overview of the statutes and guidelines most often applicable to federal drug trafficking offenses, and reports the demographics and recidivist behavior of drug trafficking offenders as a whole. Chapters Three through Seven provide detailed information about offenders as classified by the drug types studied in this report: powder cocaine, crack cocaine, heroin, marijuana, and methamphetamine. Chapter Eight concludes by reviewing the report’s findings.

Some highlights of the Commission’s study are that:

  • Over the eight-year follow-up period, one-half (50.0%) of federal drug trafficking offenders were rearrested (see bar chart). Of those drug trafficking offenders who recidivated, the median time to rearrest was 25 months.

  • In general, there were few clear distinctions among the five drug types studied. One exception is that crack cocaine offenders recidivated at the highest rate (60.8%) of any drug type. Recidivism rates for other drug types were between 43.8% and 50.0% (see table).

  • Nearly one-fourth (23.8%) of drug trafficking offenders who recidivated had assault as their most serious new charge followed by drug trafficking and public order offenses.

  • Federal drug trafficking offenders had a substantially lower recidivism rate compared to a cohort of state drug offenders released into the community in 2005 and tracked by the Bureau of Justice Statistics. Over two-thirds (76.9%) of state drug offenders released from state prison were rearrested within five years, compared to 41.9% of federal drug trafficking offenders released from prison over the same five-year period.

  • A federal drug trafficking offender’s Criminal History Category was closely associated with the likelihood of recidivism. But note that career offenders and armed career criminals recidivated at a rate lower than drug trafficking offenders classified in Criminal History Categories IV, V, and VI. (Related data and policy recommendations are discussed in the Commission's 2016 Report to the Congress on Career Offender Sentencing Enhancements.)

  • A federal drug trafficking offender’s age at time of release into the community was also closely associated with likelihood of recidivism.

February 21, 2017 in Data on sentencing, Detailed sentencing data, Drug Offense Sentencing, Federal Sentencing Guidelines, Prisons and prisoners, Purposes of Punishment and Sentencing, Reentry and community supervision | Permalink | Comments (2)

Monday, February 20, 2017

Awakening to a sleepy sentencing debate: do tired federal judges sentence more harshly?

I just came across this pair of notable papers exploring empirically whether and how less sleep might mean more punishment from federal judges:

"Sleepy Punishers Are Harsh Punishers: Daylight Saving Time and Legal Sentences" by Kyoungmin Cho, Christopher Barnes, and Cristiano Guanara

Abstract: The degree of punishment assigned to criminals is of pivotal importance for the maintenance of social order and cooperation.  Nonetheless, the amount of punishment assigned to transgressors can be affected by factors other than the content of the transgressions.  We propose that sleep deprivation in judges increases the severity of their sentences.  We took advantage of the natural quasi-manipulation of sleep deprivation during the shift to daylight saving time in the spring and analyzed archival data from judicial punishment handed out in the U.S. federal courts. The results supported our hypothesis: Judges doled out longer sentences when they were sleep deprived.

"Are Sleepy Punishers Really Harsh Punishers?: Comment" by Holger Spamann

Abstract: This comment points out four severe reservations regarding Cho et al.’s (PS 2017) finding that U.S. federal judges punish more harshly on “sleepy Mondays,” the Mondays after the start of Daylights Savings Time. First, Cho et al.'s finding pertains to only one of at least two dimensions of harshness, and the opposite result obtains in the second dimension. Second, even within the first dimension, Cho et al.'s result is statistically significant only because of a variable transformation and sample restrictions that are neither transparent in the article nor theoretically sound. Third, reanalysis of the data with superior methods reveals no significant “sleepy Monday” effect in the years 1992- 2003. Fourth, sentences were on average shorter on “sleepy Mondays” out of sample, namely in 2004-2016.

February 20, 2017 in Data on sentencing, Detailed sentencing data, Who Sentences? | Permalink | Comments (3)

Tuesday, February 14, 2017

"The American Death Penalty Decline"

The title of this post is the title of this new paper recently posted to SSRN and authored by Brandon Garrett, Alexander Jakubow and Ankur Desai. Here is the abstract:

American death sentences have both declined and become concentrated in a small group of counties. In his dissenting opinion in Glossip v. Gross in 2014, Justice Stephen Breyer argued today’s death penalty is unconstitutional, noting that from 2004 to 2006, “just 29 counties (fewer than 1% of counties in the country) accounted for approximately half of all death sentences imposed nationwide.”  That decline has become more dramatic.  Just fifty-one defendants were sentenced to death in 2015 in thirty-eight counties.  In 2016, just thirty defendants were sentenced to death in twenty-seven counties. In the mid-1990s, by way of contrast, over three hundred people were sentenced to death in as many as two hundred counties per year.

While scholars and journalists have increasingly commented on this decline and speculated as to what might be causing it, empirical research has not examined it.  This Article reports the results of statistical analysis of data hand-collected on all death sentencing, by county, for the entire modern era of capital punishment, from 1990 to 2016.  This analysis of death sentencing data from 1990 to 2016, seeks to answer the question why a few counties, but not the vast bulk of the others, still impose death sentences.  We examine state and county-level changes in murder rates, population, victim race, demography, and other characteristics that might explain shifting death sentencing patterns.

We find that death sentences are strongly associated with urban, densely populous counties.  Second, we find that death sentences are strongly associated with counties that have large black populations.  Third, we find homicide rates are related to death sentencing in three ways: contemporaneously within and between death sentencing counties, lagged within and between death sentencing counties. and that counties with more white victims of homicide have more death sentencing.  Fourth, we find that death sentencing is associated with inertia or the number of prior death sentences within a county.  These results suggest what remains of the American death penalty is quite fragile and reflects a legacy of racial bias and idiosyncratic local preferences.  We conclude by discussing the practical and legal implications of these trends for the much-diminished death penalty and for criminal justice more broadly.

February 14, 2017 in Data on sentencing, Death Penalty Reforms, Detailed sentencing data | Permalink | Comments (2)

Tuesday, January 31, 2017

"Delaying a Second Chance: The Declining Prospects for Parole on Life Sentences"

The title of this post is the title of this notable new report released today by The Sentencing Project. Here is the first part of the report's Executive Summary:

Amid growing public support for criminal justice reform, policymakers and criminal justice practitioners have begun to scale back prison sentences for low-level, nonviolent crimes. Although the results have been modest — a 5% reduction in the overall U.S. prison population between 2009 and 2015 — this shift follows almost four decades of prison expansion. But so far, criminal justice reform has largely excluded people in prison with life sentences.  This growing “lifer” population both illustrates and contributes to the persistence of mass incarceration.

Most people serving life sentences were convicted of serious crimes.  Their incarceration was intended to protect society and to provide appropriate punishment.  But many were sentenced at a time when “life with the possibility of parole” meant a significantly shorter sentence than it has become today. Many remain incarcerated even though they no longer pose a public safety risk.

Researchers have shown that continuing to incarcerate those who have “aged out” of their crime-prone years is ineffective in promoting public safety.  Long sentences are also limited in deterring future crimes given that most people do not expect to be apprehended for a crime, are not familiar with relevant legal penalties, or criminally offend with their judgment compromised by substance abuse or mental health problems.  Unnecessarily long prison terms are also costly and impede public investments in effective crime prevention, drug treatment, and other rehabilitative programs that produce healthier and safer communities.

Despite this body of criminological evidence, the number of people serving life sentences has more than quadrupled since 1984 — a faster rate of growth than the overall prison population.  Even between 2008 and 2012, as crime rates fell to historic lows and the total prison population contracted, the number of people serving life sentences grew by 12%.  By 2012, one in nine people in U.S. state and federal prisons — nearly 160,000 people — were there under life sentences.  Two factors have driven this growth: the increased imposition of life sentences, particularly those that are parole-ineligible, and an increased reluctance to grant parole to the 110,000 lifers who are eligible. MO

January 31, 2017 in Data on sentencing, Detailed sentencing data, Offender Characteristics, Offense Characteristics, Prisons and prisoners, Scope of Imprisonment, Sentences Reconsidered, Who Sentences? | Permalink | Comments (1)

Thursday, January 19, 2017

A revised empirical look at outcomes achieved by federal public defenders and court-appointed attorneys

In this post from this past summer I noted an intriguing empirical paper posted on SSRN by Yotam Shem-Tov in which the authored, after taking a deep dive into "data from all multiple defendant cases in federal courts between 2001-2014," reached the finding that "that defendants assigned a public defender in co-defendant cases had slightly worse outcomes."  A few federal public defenders let me know that the author was checking his data after receiving feedback, and the revised paper now, available here, carries a slightly different title and a significantly different key finding in its abstract:

"An Investigation of Indigent Defense Systems: Public Defenders vs. Court-Appointed Attorneys" by Yotam Shem-Tov

Abstract: To provide essential, constitutionally mandated legal services for defendants without financial means, US courts employ indigent defense systems composed of private court-appointed attorneys and public defenders’ organizations. I investigate the public defender’s causal effect on defendant sentencing outcomes relative to private court-appointed attorneys using a new “twins design” identification strategy. I argue and show empirically that in multiple defendant cases the decision of who is assigned to the public defender organization in jurisdiction X, a large urban locality, can be treated as close to a randomized experiment, which can be utilized to measure the effectiveness of court-appointed private attorneys relative to public defenders. I find that public defenders out-perform court-appointed attorneys in a range of sentencing outcomes. Employing a similar identification strategy in federal courts finds that public defenders perform at least as well if not better then court-appointed attorneys in multiple defendant cases. I provide strong evidence of selection in the assignment of attorney types to defendants in both jurisdiction X and federal courts, which makes a naıve comparison invalid and misleading.

My understanding is that the new empirical analysis now more properly accounts for the fact that public defenders typically will represent the lead defendant (and thus the one usually most culpable defendant) in multiple defendant cases, and thus a proper analysis needs to account for this critical variable.

January 19, 2017 in Data on sentencing, Detailed sentencing data, Procedure and Proof at Sentencing, Who Sentences? | Permalink | Comments (1)

Thursday, December 29, 2016

BJS releases three big reports on correctional populations throughout the United States

Via email today I received news of and links to a bunch of big data reports from the Bureau of Justice Statistics (which is part of the Office of Justice Programs at the U.S. Department of Justice). Here are the titles, links and descriptions of these notable new publications:

Correctional Populations in the United States, 2015

This report presents statistics on persons supervised by adult correctional systems in the United States at yearend 2015, including persons supervised in the community on probation or parole and those incarcerated in state or federal prison or local jail. The report describes the size and change in the total correctional population during 2015. Appendix tables provide statistics on other correctional populations and jurisdiction-level estimates of the total correctional population by correctional status and sex for selected years.

Prisoners in 2015

This report presents final counts of prisoners under the jurisdiction of state and federal correctional authorities at yearend 2015, including admissions, releases, noncitizen inmates, and inmates age 17 or younger. The report describes prisoner populations by—

  • jurisdiction
  • most serious offense
  • demographic characteristics.

Selected findings on prison capacity and prisoners held in private prisons, local jails, and the U.S. military and territories are also included. Findings are based on data from BJS's National Prisoner Statistics program, which collects data from state departments of correction and the Federal Bureau of Prisons.

Jail Inmates in 2015

This report presents information on changes in the jail inmate population between 2000 and 2015 by—

  • demographic characteristics
  • conviction status
  • average daily population
  • rated capacity of local jails
  • percent of capacity occupied.

It also includes statistics, by jurisdiction size, on changes in the number of inmates, admissions, and weekly turnover rate from 2014 to 2015. Estimates and standard errors were based on BJS's Annual Survey of Jails.

December 29, 2016 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Scope of Imprisonment, Who Sentences? | Permalink | Comments (2)

Monday, December 26, 2016

The latest data from BJS on parole and probation populations throughout the United States

Not long ago, the Bureau of Justice Statistics released this report, titled "Probation and Parole in the United States, 2015," providing the latest official data on offenders under community supervision throughout the nation. Here are some data highlights from the report:

December 26, 2016 in Criminal Sentences Alternatives, Data on sentencing, Detailed sentencing data, Reentry and community supervision | Permalink | Comments (1)

Wednesday, December 21, 2016

DPIC releases year-end report highlighting "historic declines" in use of the death penalty in 2016

NewDeathSentences1973-2016This press release from the Death Penalty Information Center, titled "Death Sentences, Executions Drop to Historic Lows in 2016," provides a summary of the DPIC's 2016 year-end report on the administration of the death penalty in the United States. Here is the text of the press release:

Death sentences, executions, and public support for capital punishment all continued historic declines in 2016.  American juries imposed the fewest death sentences in the modern era of U.S. capital punishment, since the Supreme Court declared existing death penalty statutes unconstitutional in 1972.  The expected 30 new death sentences in 2016 represent a 39 percent decline from last year’s already 40-year low of 49. The 20 executions this year marked the lowest number in a quarter century, according to a report released today by the Death Penalty Information Center (DPIC). National public opinion polls also showed support for capital punishment at a 40-year low.

“America is in the midst of a major climate change concerning capital punishment.  While there may be fits and starts and occasional steps backward, the long-term trend remains clear,” said Robert Dunham, DPIC’s Executive Director and the author of the report.  “Whether it’s concerns about innocence, costs, and discrimination, availability of life without parole as a safe alternative, or the questionable way in which states are attempting to carry out executions, the public grows increasingly uncomfortable with the death penalty each year.”

For the first time in more than 40 years, no state imposed ten or more death sentences. Only five states imposed more than one death sentence. California imposed the most (9) followed by Ohio (4), Texas (4), Alabama (3) and Florida (2).  Death sentences continued to be clustered in two percent of counties nationwide, with Los Angeles County imposing four death sentences, the most of any county. But death sentences were down 39 percent, even in those two-percent counties.

This year’s 20 executions marked a decline of more than 25 percent since last year, when there were 28 executions.  Only five states conducted executions this year, the fewest number of states to do so since 1983.  Two states -- Georgia, which had the most executions (9), and Texas, which had the second highest number (7) -- accounted for 80 percent of all executions in the U.S.  Although Georgia carried out more executions than at any other time since the 1950s, juries in that state have not imposed any new death sentences in the past two years.

State and federal courts continued to strike down outlier practices that increased the likelihood a death sentence would be imposed.  The United States Supreme Court struck down practices in Florida, Arizona, and Oklahoma that had disproportionately contributed to the number of death sentences imposed in those states.  And state courts in Florida and Delaware ruled that portions of their statutes that permitted the death penalty based upon a non-unanimous jury vote on sentencing were unconstitutional.

America’s deep divisions about capital punishment were reflected in voters’ action at the ballot box this year. Voters in California and Nebraska voted to retain the death penalty and Oklahoma voters approved a constitutional amendment regarding capital punishment.  At the same time, prosecutors in four of the 16 counties that impose the most death sentences in the U.S. were defeated by candidates who expressed personal opposition to the death penalty or pledged to reform their county’s death penalty practices.  In Kansas, pro-death penalty groups spent more than $1 million to defeat four state supreme court justices who had voted to overturn several death sentences, but voters retained all four justices.

DPIC’s review of the 20 people executed in 2016 indicated that at least 60 percent of them showed significant evidence of mental illness, brain impairment, and/or low intellectual functioning.  This suggests that, in spite of the constitutional requirement that the death penalty be reserved for the “worst of the worst” offenders, states continued to execute prisoners whose mental illness or intellectual disabilities are similar to impairments the Court has said should make a person ineligible for the death penalty.

I have reprinted above the DPIC graphic emphasizing the continued decline in the number of death sentences imposed each year because, as I have said before, I view that metric as the most significant and consequential in any serious discussion of the present status and future prospects of capital punishment throughout the US.

December 21, 2016 in Data on sentencing, Death Penalty Reforms, Detailed sentencing data, Procedure and Proof at Sentencing, Who Sentences? | Permalink | Comments (4)

Sunday, November 27, 2016

"Oregon Death Penalty: A Cost Analysis"

The title of this post is the title of this notable research report released earlier this month.  This press release from Lewis & Clark Law School provides helpful background on the report and its findings. Here are excerpts:

A new report by Lewis & Clark Law School and Seattle University offers an unprecedented financial picture of the previously uncalculated cost of capital punishment in Oregon. “Oregon Death Penalty: A Cost Analysis” shows that the costs for aggravated murder cases that result in death sentences range, on average, 3.5 to 4 times more expensive per case when compared to similar non-death penalty cases.

Lewis & Clark Law Professor Aliza Kaplan spearheaded the research effort, fueled by the fact that there was no data to answer questions about the cost of capital punishment in Oregon. Kaplan approached co-author Peter A. Collins, PhD of Seattle University’s Criminal Justice Department, to complement her legal analysis with best-in-class quantitative analysis methods, following his similar 2015 report on death-penalty cost analysis for the state of Washington.

Looking at cost data from the Oregon Department of Corrections (DOC), the Oregon Department of Justice (DOJ), and the Oregon Office Public Defense Services (OPDS) among other sources, the report also examines the role that the lengthiness of death penalty cases plays in their total costs. These cases stretch on for decades due to the constitutional and statutory requirements of appeals and reconsiderations, which increases the net litigation costs for all parties.

The report, which took more than 18 months to compile, also looks at the use of the death penalty in Oregon, which voters did away with in 1964, but reinstated two decades later. Since 1984, 62 individuals have been convicted and sentenced to death. Of those 62, twenty-eight of them are no longer on death row. Just two of these cases have 1 resulted in death (both individuals dropped their appeals and “volunteered” to be executed), four people died of natural causes while in prison, and 22 people, or roughly 79%, have had their sentences reduced.

Offering common ground for policymakers and citizens of Oregon to examine capital punishment, the report is part of a growing trend to bring better data to the work of crafting more sound public policy. For Kaplan, the report is about increasing transparency through better data. “The decision makers, those involved in the criminal justice system, everyone, deserves to know how much we are currently spending on the death penalty, so that when stakeholders, citizens and policy-makers make these decisions, they have as much information as possible to decide what is best for Oregon,” said Kaplan.

According to Dr. Peter Collins, “There are several important takeaways from this research for Oregonians. First, the evidence clearly shows that aggravated murder cases that involve the death penalty are at least three-and-a-half to four-times more expensive than aggravated murder cases that do not involve the death penalty. Second, although the death penalty is not being pursued as frequently as in the past, the average costs when it is have markedly increased. Last, it is ultimately a futile endeavor, as the vast majority of death penalty sentences are decreased to life without parole in post conviction appeals.”

Six law students at Lewis & Clark provided key assistance in producing the report, conducting extensive legal research and field interviews with professionals throughout the criminal justice system. Third-year law student and co-author of the study, Venetia Mayhew, was involved in the project since its first day. “Professor Kaplan provided me with a remarkable opportunity to delve deep into Oregon’s death penalty system and to understand the laborious and costly nature of its processes. I was most struck by the human cost it imposes on all those who participate,” said Venetia Mayhew, JD ’17, who began her work on the analysis in her first year as a Lewis & Clark law student.

November 27, 2016 in Data on sentencing, Death Penalty Reforms, Detailed sentencing data, Purposes of Punishment and Sentencing, Who Sentences? | Permalink | Comments (3)

Thursday, October 27, 2016

Significant new report calls for closing all traditional youth prisons due to their inefficacy

This recent item from the Harvard Gazette, headlined "Youth justice study finds prison counterproductive: New report documents urgent need to replace youth prisons with rehabilitation-focused alternatives," spotlights a significant new report concerning the way juvenile offenders are punished. Here are excerpts:

A new report, published by Harvard Kennedy School’s Program in Criminal Justice Policy and Management (PCJ) and the National Institute of Justice (NIJ), documents ineffectiveness, endemic abuses, and high costs in youth prisons throughout the country.  The report systematically reviews recent research in developmental psychology and widespread reports of abuse to conclude that the youth prison model should be replaced with a continuum of community-based programs and, for the few youth who require secure confinement, smaller homelike facilities that prioritize age-appropriate rehabilitation.

The authors, who are leading youth justice researchers and former youth correctional administrators, find that the current youth prison model, which emphasizes confinement and control, often exacerbates youth trauma and inhibits positive growth while failing to address public safety.  Rather, the paper argues, programs work best when youths are in their home communities with rehabilitative programs or in smaller, homelike facilities that promote opportunities for healthy decision-making and development. Corrections agencies should provide a range of options depending on the individual’s needs, from smaller secure facilities to noncustodial programs.

Annual youth imprisonment costs are approximately $150,000 per individual, yet recidivism rates remain close to 70 percent. The report examines the experiences of several states that have pursued alternative models and finds community-based approaches can reduce recidivism, control costs, and promote public safety.

“Youth in trouble need guidance, education, and support, not incarceration in harmful and ineffective youth prisons,” said PCJ Senior Fellow Vincent Schiraldi, a co-author of the report. Previously, Schiraldi directed juvenile corrections in Washington, D.C., and served as commissioner of probation in New York City.  “We now know from research and on-the-ground experience that youth prisons are not designed to best promote youth rehabilitation.  This report offers concrete alternatives for policymakers across the country to maintain public safety, hold young people accountable, and turn their lives around.”

“Juvenile-justice systems must have the clear purpose of giving each youth the tools he or she needs to get on the right path to a successful adulthood and to reintegrate into the community,” said Patrick McCarthy, president and chief executive officer of the Annie E. Casey Foundation and a co-author of the report.  Like Schiraldi, McCarthy is a former director of youth corrections — in his case, in Delaware.  “By closing traditional youth prisons and leveraging increased political will to reform our country’s dependence on incarceration, states can use the savings to begin implementing a new, more effective approach to serving young people.”

This report, titled “The Future of Youth Justice: A Community-Based Alternative to the Youth Prison Model,” is available in full at this link.  And here is a key paragraph from its opening pages:

Whether the benefits and costs of youth prisons are weighed on a scale of public dollars, community safety, or young people’s futures, they are damaging the very people they are supposed to help and have been for generations. It is difficult to find an area of U.S. policy where the benefits and costs are more out of balance, where the evidence of failure is clearer, or where we know with more clarity what we should be doing differently.

October 27, 2016 in Data on sentencing, Detailed sentencing data, Offender Characteristics, Prisons and prisoners, Scope of Imprisonment | Permalink | Comments (1)

Monday, October 17, 2016

Thoughtful look into fairness/bias concerns with risk-assessment instruments like COMPAS

A group of Stanford professors and students have this thoughtful new Washington Post commentary headlined "A computer program used for bail and sentencing decisions was labeled racist. It’s actually not that clear." The piece is a must-read for everyone concerned about risk-assessment technologies (which should be everyone).  Here are excerpts:

This past summer, a heated debate broke out about a tool used in courts across the country to help make bail and sentencing decisions. It’s a controversy that touches on some of the big criminal justice questions facing our society. And it all turns on an algorithm.

The algorithm, called COMPAS, is used nationwide to decide whether defendants awaiting trial are too dangerous to be released on bail. In May, the investigative news organization ProPublica claimed that COMPAS is biased against black defendants. Northpointe, the Michigan-based company that created the tool, released its own report questioning ProPublica’s analysis. ProPublica rebutted the rebuttal, academic researchers entered the fray, this newspaper’s Wonkblog weighed in, and even the Wisconsin Supreme Court cited the controversy in its recent ruling that upheld the use of COMPAS in sentencing.

It’s easy to get lost in the often technical back-and-forth between ProPublica and Northpointe, but at the heart of their disagreement is a subtle ethical question: What does it mean for an algorithm to be fair? Surprisingly, there is a mathematical limit to how fair any algorithm — or human decision-maker — can ever be.

The COMPAS tool assigns defendants scores from 1 to 10 that indicate how likely they are to reoffend based on more than 100 factors, including age, sex and criminal history. Notably, race is not used. These scores profoundly affect defendants’ lives: defendants who are defined as medium or high risk, with scores of 5-10, are more likely to be detained while awaiting trial than are low-risk defendants, with scores of 1-4.

We reanalyzed data collected by ProPublica on about 5,000 defendants assigned COMPAS scores in Broward County, Fla. (See the end of the post, after our names, for more technical details on our analysis.) For these cases, we find that scores are highly predictive of reoffending. Defendants assigned the highest risk score reoffended at almost four times the rate as those assigned the lowest score (81 percent vs. 22 percent).

Northpointe contends they are indeed fair because scores mean essentially the same thing regardless of the defendant’s race. For example, among defendants who scored a seven on the COMPAS scale, 60 percent of white defendants reoffended, which is nearly identical to the 61 percent of black defendants who reoffended. Consequently, Northpointe argues, when judges see a defendant’s risk score, they need not consider the defendant’s race when interpreting it....

But ProPublica points out that among defendants who ultimately did not reoffend, blacks were more than twice as likely as whites to be classified as medium or high risk (42 percent vs. 22 percent). Even though these defendants did not go on to commit a crime, they are nonetheless subjected to harsher treatment by the courts. ProPublica argues that a fair algorithm cannot make these serious errors more frequently for one race group than for another.

Here’s the problem: it’s actually impossible for a risk score to satisfy both fairness criteria at the same time.... If Northpointe’s definition of fairness holds, and if the recidivism rate for black defendants is higher than for whites, the imbalance ProPublica highlighted will always occur.

It’s hard to call a rule equitable if it does not meet Northpointe’s notion of fairness. A risk score of seven for black defendants should mean the same thing as a score of seven for white defendants. Imagine if that were not so, and we systematically assigned whites higher risk scores than equally risky black defendants with the goal of mitigating ProPublica’s criticism. We would consider that a violation of the fundamental tenet of equal treatment.

But we should not disregard ProPublica’s findings as an unfortunate but inevitable outcome. To the contrary, since classification errors here disproportionately affect black defendants, we have an obligation to explore alternative policies. For example, rather than using risk scores to determine which defendants must pay money bail, jurisdictions might consider ending bail requirements altogether — shifting to, say, electronic monitoring so that no one is unnecessarily jailed.

COMPAS may still be biased, but we can’t tell. Northpointe has refused to disclose the details of its proprietary algorithm, making it impossible to fully assess the extent to which it may be unfair, however inadvertently. That’s understandable: Northpointe needs to protect its bottom line. But it raises questions about relying on for-profit companies to develop risk assessment tools.

Moreover, rearrest, which the COMPAS algorithm is designed to predict, may be a biased measure of public safety. Because of heavier policing in predominantly black neighborhoods, or bias in the decision to make an arrest, blacks may be arrested more often than whites who commit the same offense.

Algorithms have the potential to dramatically improve the efficiency and equity of consequential decisions, but their use also prompts complex ethical and scientific questions. The solution is not to eliminate statistical risk assessments. The problems we discuss apply equally to human decision-makers, and humans are additionally biased in ways that machines are not. We must continue to investigate and debate these issues as algorithms play an increasingly prominent role in the criminal justice system.

Some (of many) prior related posts on use of risk-assessment technologies:

October 17, 2016 in Data on sentencing, Detailed sentencing data, Procedure and Proof at Sentencing, Purposes of Punishment and Sentencing, Race, Class, and Gender, Technocorrections, Who Sentences? | Permalink | Comments (0)

Thursday, October 13, 2016

Fair Punishment Project releases second part of report on small number of US counties still actively utilizing the death penalty

In this post earlier this year, I noted the significant new initiative emerging from Harvard Law School's Charles Hamilton Houston for Race & Justice and its Criminal Justice Institute called the Fair Punishment Project (FPP).   And in this post a couple of months ago, I highlighted the new big project and first part of a report from the the FPP providing an in-depth look at how the death penalty is operating in the handful of counties still actively using it.  The second part of this report has now been released under the title "Too Broken to Fix, Part II: An In-depth Look at America’s Outlier Death Penalty Counties," and it is available at this link. Here is its introduction:  

As we noted in Part I of this report, the death penalty in America is dying.

In 2015, juries only returned 49 death sentences — the fewest number since the death penalty was reinstated in 1976.  Of the 31 states that legally retain the death penalty, only 14 — or less than half — imposed a single death sentence in 2015.  When we look at the county level, the large-scale abandonment of the death penalty in the country becomes even more apparent.  Of the 3,143 county or county equivalents in the United States, only 33 counties — or one percent — imposed a death sentence in 2015. Just 16 — or one half of one percent — imposed five or more death sentences between 2010 and 2015.  Among these outliers, six are in Alabama (Jefferson and Mobile) and Florida (Duval, Hillsborough, Miami-Dade, and Pinellas)—the only two states that currently permit non-unanimous death verdicts.  Of the remaining 10 counties, five are located the in highly-populated Southern California region (Kern, Los Angeles, Orange, Riverside, and San Bernardino). The others include Caddo Parish (LA), Clark (NV), Dallas (TX), Harris (TX), and Maricopa (AZ). As Justice Stephen Breyer noted in his 2015 dissent in Glossip v. Gross, “the number of active death penalty counties is small and getting smaller.”

In this two-part report, we have endeavored to figure out what makes these 16 counties different by examining how capital punishment operates on the ground in these outlier death-sentencing counties. In Part II, we highlight Dallas (TX), Jefferson (AL), San Bernardino (CA), Los Angeles (CA), Orange (CA), Miami-Dade (FL), Hillsborough (FL), and Pinellas (FL) counties.

Our review of these counties, like the places profiled in Part I, reveals that these counties frequently share at least three systemic deficiencies: a history of overzealous prosecutions, inadequate defense lawyering, and a pattern of racial bias and exclusion. These structural failings regularly produce two types of unjust outcomes which disproportionately impact people of color: the wrongful conviction of innocent people, and the excessive punishment of persons who are young or suffer from severe mental illnesses, brain damage, trauma, and intellectual disabilities.

This is what capital punishment in America looks like today. While the vast majority of counties have abandoned the practice altogether, what remains is the culmination of one systemic deficiency layered atop another.  Those who receive death sentences do not represent the so-called “worst of the worst.”  Rather, they live in counties with overzealous and often reckless prosecutors, are frequently deprived access to competent and effective representation, and are affected by systemic racial bias.  These individuals are often young, and many have significant mental impairments. Some are likely innocent.  This pattern offers further proof that, whatever the death penalty has been in the past, today it is both cruel and unusual, and therefore unconstitutional under the Eighth Amendment.

Prior related posts:

October 13, 2016 in Death Penalty Reforms, Detailed sentencing data, Procedure and Proof at Sentencing, Race, Class, and Gender, Who Sentences? | Permalink | Comments (3)

Monday, September 26, 2016

FBI releases "official" 2015 US crime statistics showing increase in violent crime (especially murderes) and decreased property crime

As reported in this official FBI press release, "[a]fter two years of decline, the estimated number of violent crimes in the nation increased 3.9 percent in 2015 when compared with 2014 data, according to FBI figures released today. Property crimes dropped 2.6 percent, marking the 13th straight year the collective estimates for these offenses declined." This short FBI report on its latest data provides these additional particulars and helpful context:

Today, the FBI released its annual compilation of crimes reported to its Uniform Crime Reporting (UCR) Program by law enforcement agencies from around the nation. Crime in the United States, 2015 reveals a 3.9 percent increase in the estimated number of violent crimes and a 2.6 percent decrease in the estimated number of property crimes last year when compared to 2014 data.

According to the report, there were an estimated 1,197,704 violent crimes committed around the nation.  While that was an increase from 2014 figures, the 2015 violent crime total was 0.7 percent lower than the 2011 level and 16.5 percent below the 2006 level.

Among some of the other statistics contained in Crime in the United States, 2015:

  • The estimated number of murders in the nation was 15,696. [This is a roughly 11% increase from 2014.]

  • During the year, there were an estimated 90,185 rapes. (This figure currently reflects UCR’s legacy definition.....) [This is a roughly 6% increase from 2014.]

  • There were an estimated 327,374 robberies nationwide, which accounted for an estimated $390 million in losses (average dollar value of stolen property per reported robbery was $1,190).

  • Firearms were used in 71.5 percent of the nation’s murders, 40.8 percent of robberies, and 24.2 percent of aggravated assaults.

  • Property crimes resulted in losses estimated at $14.3 billion. The total value of reported stolen property (i.e., currency, jewelry, motor vehicles, electronics, firearms) was $12,420,364,454.

Like all detailed and intricate numbers about crime and punishment, these latest data can (and surely will) be spun in all sorts of ways.  For some early examples of the spin, here are some early commentaries about the data:

From Crime & Consequences here, "Complacency Mongers, Start Your Engines!"

From the Daily Beast here, "Violent Crime Is Up, but Trump Is Still Wrong"

From the Huffington Post here, "2015 Was One Of The Safest Years In The Past 2 Decades, According To FBI Crime Stats"

September 26, 2016 in Data on sentencing, Detailed sentencing data, National and State Crime Data | Permalink | Comments (5)

Friday, September 23, 2016

Great new US Sentencing Commission report on "simple possession" federal drug cases raises array of hard follow-up questions

Simplepossession_coverI find crime and punishment data so interesting and so important in large part because (1) even seemingly basic and simple data often can only be fully understood after one takes time to examine closely the backstories that surround that data, and (2) only if and when a researcher or advocate has deep understanding of data can that person even start to appreciate all the challenging policy and practical questions that important data implicate.  These realities are on full display in the context of an interesting and important new report released this week by the US Sentencing Commission titled "Weighing the Charges: Simple Possession of Drugs in the Federal Criminal Justice System." Here is the introduction to the short report, which explains the notable backstories concerning a dramatic recent change in the number of federal "simple possession" cases:

The simple possession of illegal drugs is a criminal offense under federal law and in many state jurisdictions. The offense occurs “when someone has on his or her person, or available for his or her use, a small amount of an illegal substance for the purpose of consuming or using it but without the intent to sell or give it to anyone else.”

Simple drug possession is a misdemeanor under federal law which provides that an offender may be sentenced to a term of imprisonment of not more than one year, fined a minimum of $1,000, or both. However, if an offender is convicted of simple possession after a prior drug related offense has become final, the offender can be charged with a felony simple possession offense.

The number of federal offenders whose most serious offense was simple drug possession increased nearly 400 percent during the six-year period between fiscal years 2008 and 2013. A change of this magnitude over a relatively short period of time triggered further investigation into these cases using data on offender and offense characteristics routinely collected by the United States Sentencing Commission (“the Commission”), as well as additional data collected specifically for this project.

At first, this dramatic increase in the number of offenders sentenced for the simple possession of drugs seems to suggest a substantially increased focus on this offense by federal law enforcement personnel. Further analysis, however, does not support such a conclusion. A closer inspection of the data demonstrates that this increase is almost entirely attributable to a single drug type — marijuana — and to offenders who were arrested at or near the U.S./Mexico border (a group almost entirely composed of offenders from the District of Arizona). For simple possession of marijuana offenders arrested at locations other than the U.S./Mexico border, the median quantity of marijuana involved in the offense was 5.2 grams (0.2 ounces).  In contrast, the offense conduct of simple possession of marijuana offenders arrested at that border involved a median quantity of 22,000 grams (48.5 pounds or 776.0 ounces) — a quantity that appears in excess of a personal use quantity.

In other words, the USSC noticed data showing a huge increase in the charging of misdemeanor federal drug crimes, which at first might suggest a curious new commitment by federal prosecutors to pursue low-level drug offenders. But, upon closer examination, the USSC discovers that what is really going on is that a whole lot of (low-level?) drug traffickers (mules?) found with huge quantities of marijuana are having their cases prosecuted through "simple possession" charges even though that label hardly seems like a factually fitting description of their drug crimes.

I am extraordinarily pleased to see the USSC detailing and explaining this interesting new data trend, and I am extraordinarily interested to hear from readers as to whether they think federal prosecutors in border regions ought to be praised or pilloried for their new misdemeanor approach to dealing with marijuana offenders arrested at the border with an average of 50 pounds of mary jane. This USSC report not only documents one tangible way that state marijuana reforms would seem to be having a profound impact on how the federal government is now waging the so-called "war on weed," but it also prompts a lot of hard questions about whether the new behaviors by federal drug prosecutors are appropriate given the absence of any formal changes to federal drug laws.

September 23, 2016 in Data on sentencing, Detailed sentencing data, Drug Offense Sentencing, Federal Sentencing Guidelines, Marijuana Legalization in the States, Pot Prohibition Issues, Who Sentences? | Permalink | Comments (3)

Sunday, September 04, 2016

"The 'Cost of Crime' and Benefit-Cost Analysis of Criminal Justice Policy: Understanding and Improving Upon the State-of-The-Art"

The title of this post is the title of this notable new paper authored by Mark Cohen and available via SSRN. Here is the abstract:

The use of benefit-cost analyses by criminal justice researchers has slowly been increasing over the past 30 years. While still in its infancy, benefit-cost analyses of criminal justice policies have recently moved from the academic arena to actual use by policy makers.  The growing use of benefit-cost analysis in crime policy tends to be lauded by economists; however, criminologists and legal scholars are less than unanimous in their views.  A recent issue of Criminology and Public Policy on the “Role of the Cost-of-Crime Literature,” highlights this controversy.

Two main themes can be distilled from critiques of the literature: first, the considerable uncertainty that exists in cost and benefit estimates; and second, the fact that important social costs are not being taken into account in current models.  A related critique is that current methodologies to estimate the cost of crime are affected by income; bringing with it a concern that criminal justice policies based on a benefit-cost analysis will favor the rich.

This research note addresses these broad questions about the role of benefit-cost analysis in the criminal justice policy arena and attempts to clear up some misunderstandings on the methodologies used to estimate the cost of crime.  I also highlight some of the most important gaps in the literature for those interested in helping to improve the state-of-the-art in estimating the “cost of crime.”  Perhaps equally important, I hope to demystify benefit-cost analysis and unmask it for what it really is — an important tool that can help policy makers systematically and transparently assess and compare options with the goal of making better policy decisions.

September 4, 2016 in Detailed sentencing data, National and State Crime Data, Purposes of Punishment and Sentencing | Permalink | Comments (0)

Wednesday, August 24, 2016

Important "Real Clear" debate explores whether Texas "smart on crime" reforms have really been successful

A series of dueling posts over at the Real Clear Policy blog has been engaging with crime and punishment data from Texas to provide different views on whether so-called "smart on crime" reforms in the Lone Star State have proven truly effective at reducing both crime and imprisonment.  The discussion is too intricate to summarize here, so I encourage readers interested in this important debate to check out these post in order:

August 24, 2016 in Data on sentencing, Detailed sentencing data, National and State Crime Data, Prisons and prisoners, State Sentencing Guidelines | Permalink | Comments (0)

Tuesday, August 23, 2016

New Fair Punishment Project report takes close look at small number of US counties still actively utilizing the death penalty

FairJustIn this post earlier this year, I noted the new initiative emerging from Harvard Law School's Charles Hamilton Houston for Race & Justice and its Criminal Justice Institute called the Fair Punishment Project (FPP).   I received an email this morning highlighting a new big project and report from the the FPP.  Here are excerpts from the email:

Today [FPP] released a new report offering an in-depth look at how the death penalty is operating in the small handful of counties across the country that are still using it.  Of the 3,143 county or county equivalents in the United States, only 16—or one half of one percent—imposed five or more death sentences between 2010 and 2015.  Part I of the report, titled Too Broken to Fix: An In-depth Look at America’s Outlier Death Penalty Counties examined 10 years of court opinions and records from eight of these 16 “outlier counties,” including Caddo Parish (LA), Clark (NV), Duval (FL), Harris (TX), Maricopa (AZ), Mobile (AL), Kern (CA) and Riverside (CA). The report also analyzed all of the new death sentences handed down in these counties since 2010....

The report notes that these “outlier counties” are plagued by persistent problems of overzealous prosecutors, ineffective defense lawyers, and racial bias. Researchers found that the impact of these systemic problems included the conviction of innocent people, and the excessively harsh punishment of people with significant impairments.  The report notes that many of the defendants appear to have one or more impairments that are on par with, or worse than, those that the U.S. Supreme Court has said should categorically exempt individuals from execution due to lessened culpability.  The Court previously found that individuals with intellectual disabilities (Atkins v. Virginia, 2002) and juveniles under the age of 18 (Roper v. Simmons, 2005) should not be subject to the death penalty under the Eighth Amendment.

In conducting its analysis, we reviewed more than 200 direct appeals opinions handed down between 2006 and 2015 in these eight counties. We found:

  • Sixty percent of cases involved defendants with significant mental impairments or other forms of mitigation.
  • Eighteen percent of cases involved a defendant who was under the age of 21 at the time of the offense. In Riverside County, 16 percent of the defendants were age 18 at the time of the offense.
  • Forty-four percent of cases involved a defendant who had an intellectual disability, brain damage, or severe mental illness. In four of the counties, half or more of the defendants had mental impairments: Maricopa (62 percent), Mobile (60 percent), Caddo Parish (57 percent), and Kern (50 percent).
  • Approximately one in seven cases involved a finding of prosecutorial misconduct. Maricopa and Clark counties had misconduct in 21 percent and 47 percent of cases respectively.
  • Bad lawyering was a persistent problem across all of the counties. In most of the counties, the average mitigation presentation at the penalty phase of the trial, in which the defense lawyer is supposed to present all of the evidence showing that the defendant’s life should be spared -- including testimony from mental health and other experts, lasted approximately one day. While this is just one data point for determining the quality of legal representation, this finding reveals appalling inadequacies. In Duval County, Florida, the entire penalty phase of the trial and the jury verdict often came in the same day.
  • A relatively small group of defense lawyers represented a substantial number of the individuals who ended up on death row. In Kern County, one lawyer represented half of the individuals who ended up on death row between 2010 and 2015.

Additional findings:

  • Five of the eight counties had at least one person exonerated from death row since 1976. Harris County has had three death row exonerations, and Maricopa has had five.
  • Out of all of the death sentences obtained in these counties between 2010 and 2015, 41 percent were given to African-American defendants, and 69 percent were given to people of color.  In Duval, 87 percent of defendants were Black in this period. In Harris, 100 percent of the defendants who were newly sentenced to death since November 2004 have been people of color.
  • The race of the victim is also a significant factor in who is sentenced to death in many of these counties. In Mobile County, 67 percent of the Black defendants, and 88% of all defendants, who were sentenced to death were convicted of killing white victims. In Clark County, 71 percent of all of the victims were white in cases resulting in a death sentence. The report noted just three white defendants sentenced to death for killing Black victims between 2010 and 2015. One of those cases was from Riverside, and in that case the defendant was also convicted of killing two additional white victims. The two other cases were from Duval.
  • Five of the 16 “outlier counties” are from Florida and Alabama, the only two states that currently allow non-unanimous jury verdicts.  In Duval, 88 percent of the decisions in the review period were non-unanimous, and in Mobile the figure was 80 percent. 

Part II of this report, which will be released in September, will look at the remaining eight “outlier counties,” including: Dallas (TX), Jefferson (AL), Pinellas (FL), Miami-Dade (FL), Hillsborough (FL), Los Angeles (CA), San Bernardino (CA), and Orange (CA).

August 23, 2016 in Data on sentencing, Death Penalty Reforms, Detailed sentencing data, Offender Characteristics, Procedure and Proof at Sentencing, Race, Class, and Gender, Who Sentences? | Permalink | Comments (1)

Sunday, August 21, 2016

Some surprising racial realities to discover when taking a deep dive into modern mass incarceration data

U.S._incarceration_rates_1925_onwardsA couple of folks have pointed me to this recent interesting analysis at Wonkblog by Keith Humphreys under the headline "Black incarceration hasn’t been this low in a generation." Here are some of the data and discussion that explain the headline (with links from the original):

The African American imprisonment rate has been declining for many years. Indeed, the likelihood of African American men and women being in prison today is lower than it was a generation ago ... [because the] rate of black male incarceration in the U.S. has declined by 23 percent from a recent peak in 2001 [and the] rate of incarcerated black women has decreased 49 percent since the recent peak of 1999....

In the 1990s, the explosive growth in imprisonment that began in the mid-1970s was slowing but still underway, affecting people of all races but African Americans worst of all.  But around the turn of the millennium, the African American imprisonment rate began declining year after year....

At the end of 2014, the African American male imprisonment rate had dropped to a level not seen since early 1993. The change for African American women is even more marked, with the 2014 imprisonment rate being the lowest point in the quarter-century of data available. It can’t be overemphasized that these are trends unique to blacks rather than being part of a broader pattern of de-incarceration: The white imprisonment rate has been rising rather than falling.

A 23 percent decline in the black male imprisonment rate and a 49 percent decline in the black female imprisonment rate would seem to warrant some serious attention. But if you point out to the average person or even a seasoned criminologist that the United States is at a more than 20-year low in the black incarceration rate, you are likely to be met with stunned silence.

These data should not be all that surprising for those who realize that the years from 1970 to 2000 marked the modern period with the most significant increase in incarceration rates for all Americans and particularly for African Americans.  Since 2000, the overall US prison population has not grown much, and overall prison populations and the rate of incarceration has even turned downward in recent years.  I believe that, during this more recent period of flat or declining prison growth, the emphasis in long prison terms less for drug offenders than for violent/sexual offenders has contributing to altering the racial mix of prison populations (perhaps epsecially in big states like California and Texas that have made big cuts in their prison populations).

That all said, these data should not obscure the reality that incarceration rates for black males remain extraordinarily high both in absolute and in relative terms throughout the United States.  Moreover, digging into state-by-state incarceration data highlights that some perhaps unexpected states rise to the top of an accounting of the rate and relative levels of minority incarceration.  A few months ago (as noted here), The Sentencing Project released this interesting report providing state-by-state analyses of the racial data for state prison populations, and here were some of the report's "Key Findings":

August 21, 2016 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Race, Class, and Gender, Scope of Imprisonment, Who Sentences? | Permalink | Comments (4)

Thursday, August 11, 2016

"Public Defenders vs. Private Court Appointed Attorneys: An Investigation of Indigent Defense Systems"

The title of this post is the title of this intriguing empirical paper recently posted on SSRN authored by Yotam Shem-Tov. Here is the abstract (with one line emphasized by me):

Individuals facing criminal charges in the U.S. have a constitutional right to attorney representation.  If they cannot afford one, the court is required to appoint and finance a legal counsel. Indigent defense systems are usually composed of private court appointed attorneys and/or a public defenders’ organization.  I investigate the causal effect of being assigned a public defender as oppose to a private court appointed attorney on defendants’ trial outcomes using a new “twins design” identification strategy.

I argue and show empirically that in co-defendant cases, the decision of who is assigned to the public defender organization can be treated as close to a randomized experiment, which can be exploited to measure the effectiveness of court appointed private attorneys relative to public defenders.  Using data from all multiple defendant cases in federal courts between 2001-2014, I find that defendants assigned a public defender in co-defendant cases had slightly worse outcomes: a higher probability of being convicted, an average of three months longer expected prison sentence, longer court proceedings and a higher probability of reaching a plea bargain.  However, there is large heterogeneity in public defender effectiveness across federal districts ranging from a 13.8 month longer prison term to a 16.1 month shorter prison term.

I lack the empirical skills to question or assess the new “twins design” identification strategy used in this paper, and the full paper is full of challenging empirical jargon like "The main regression specications is Yit = B . PDi + aj(I) + X'it􀀀 + Eit".  That said, my first reaction is to be quite suspicious of these findings/conclusions due to (1) my own experiences and high regard for the work of nearly all federal public defenders, and (2) my own normative instinct that, at least for a significant number of federal defendants, experiencing "longer court proceedings and a higher probability of reaching a plea bargain" is not at all a marker of a "worse outcome."

UPDATE months later: As noted in this post, the author revised this paper based on feedback and a new data analysis and reached a significantly different finding: "I find that public defenders out-perform court-appointed attorneys in a range of sentencing outcomes."

August 11, 2016 in Data on sentencing, Detailed sentencing data, Procedure and Proof at Sentencing, Who Sentences? | Permalink | Comments (11)

Wednesday, August 03, 2016

Does the weather and MLB baseball impact federal sentencing outcomes more than racial factors?

The seemingly somewhat kooky question in the title of this post is prompted by this seemingly somewhat kooky empirical paper now available via SSRN and authored by a group of data researchers and titled "Events Unrelated to Crime Predict Criminal Sentence Length." Here is the paper's abstract (with a key sentence emphasized to explain my post title query):

In United States District Courts for federal criminal cases, prison sentence length guidelines are established by the severity of the crime and the criminal history of the defendant.  In this paper, we investigate the sentence length determined by the trial judge, relative to this sentencing guideline.  Our goal is to create a prediction model of sentencing length and include events unrelated to crime, namely weather and sports outcomes, to determine if these unrelated events are predictive of sentencing decisions and evaluate the importance weights of these unrelated events in explaining rulings.

We find that while several appropriate features predict sentence length, such as details of the crime committed, other features seemingly unrelated, including daily temperature, baseball game scores, and location of trial, are predictive as well.  Unrelated events were, surprisingly, more predictive than race, which did not predict sentencing length relative to the guidelines.  This is consistent with recent research on racial disparities in sentencing that highlights the role of prosecutors in making charges that influence the maximum and minimum recommended sentence.  Finally, we attribute the predictive importance of date to the 2005 U.S. Supreme Court case, United States v. Booker, after which sentence length more frequently fell near the guideline minimum and the range of minimum and maximum sentences became more extreme.

Based on a quick scan of the paper, I came to the conclusion that one would need to have a pretty sophisticated understanding of both federal sentencing patterns and empirical methods to assess the soundness of the analysis here.  Still, the paper's penultimate paragraph reinforces that this analysis led to some notable conclusions (with my emphasis again added):

A justice system reasonably aspires to be consistent in the application of law across cases and to account for the particulars of a case. Our goal was to create a prediction model of criminal sentence lengths that accounts for non-judicial factors such as weather and sports events among the feature set. The feature weights offer a natural metric to evaluate the importance of these features unrelated to crime relative to case-specific factors. Using a Random Forest, we found several expected crime related features appearing within the top 10% most important features. However, we also found defendant characteristics (unrelated to the crime), sport game outcomes, weather, and location features all predictive of sentence length as well, and these features were, surprisingly, more predictive than the defendant’s race. Further investigating this predictive ability would be of interest to those studying the criminal justice system.

August 3, 2016 in Booker in district courts, Data on sentencing, Detailed sentencing data, Offense Characteristics, Who Sentences? | Permalink | Comments (1)

Thursday, July 28, 2016

US Sentencing Commission releases big new report urging reform of career offender enhancements

As detailed in this official press release, the US Sentencing Commission today released a big new report (running over 100 pages!) under the title "Report to the Congress: Career Offender Sentencing Enhancements." Here is how the press release summarizes this important new release from the USSC:

The United States Sentencing Commission (“Commission”) issued a Report to the Congress: Career Offender Sentencing Enhancements, analyzing career offenders’ prior criminal history, incarceration terms and recidivism rates.

Chief Judge Patti B. Saris, Chair of the Commission, stated, “The Commission’s research shows that there are important differences between violent career offenders and drug trafficking career offenders. Based on these findings, Congress should amend the statutory criteria such that career offender status would not be based solely on drug trafficking offenses.”

Currently, a defendant qualifies as a career offender if he or she: 1) is convicted of an offense that is either a crime of violence or a controlled substance offense; and 2) has at least two prior felony convictions.  Career offenders face longer incarceration terms, receiving an average sentence of more than 12 years (147 months).  As a result of these longer sentences, career offenders now account for more than 11 percent of the total federal prison population.  Yet, career offenders are increasingly receiving sentences below the federal sentencing guideline range, often at the request of the government.  The research also shows that, compared to “drug trafficking only” offenders, violent career offenders generally have a more serious and extensive criminal history, recidivate at a higher rate, and are more likely to commit another violent offense in the future.  In fiscal year 2014, 45% of “drug trafficking only” offenders received sentences that were reduced at the government’s request.

In fiscal year 2014, nearly three-quarters (74.1%) of career offenders were convicted of a drug trafficking offense. Drug trafficking offenders often face higher statutory maximum penalties, including life imprisonment.  These offenders were also more likely to receive a sentence below the federal sentencing guideline range.

Earlier this year, the Commission voted unanimously to amend the definition of “crime of violence” in the federal sentencing guidelines, with an effective date of August 1, 2016.  Chair Saris added, “Based on the report’s findings and recommendations, Congress should adopt a new, single definition of ‘crime of violence’ that is consistent with the Commission’s revised approach.”

July 28, 2016 in Data on sentencing, Detailed sentencing data, Offender Characteristics, Purposes of Punishment and Sentencing, Scope of Imprisonment, Who Sentences? | Permalink | Comments (1)

Thursday, June 30, 2016

New report highlights huge role of a handful of local prosecutors on the size of death rows

20160628_FPP-ShareableThis notable new report from Harvard Law School’s Fair Punishment Project highlights the consequential role of just a handful of local prosecutors on the modern US death penalty.  The report, titled "America’s Top Five Deadliest Prosecutors: How Overzealous Personalities Drive The Death Penalty," gets started this way (with footnotes removed):

Last year, a journalist asked Dale Cox, then the District Attorney of Caddo Parish, Louisiana, about the wisdom of the death penalty in light of the recent exoneration of Glenn Ford, a man who spent thirty years on death row for a crime that he did not commit.  Cox told the reporter: “I think we need to kill more people.” “Revenge,” he said, “brings to us a visceral satisfaction.”  Between 2010 and 2015, Cox alone secured one-third of Louisiana’s death sentences.

Cox’s disproportionate use of the death penalty illustrates a point that Justice Stephen Breyer recently made. “It is now unusual to find capital punishment in the United States,” Breyer wrote, because “capital prosecutions are being pursued in only a few isolated counties.”  There are more than 3,100 counties, 2,400 head prosecutors, and thousands of line prosecutors in America — yet only a tiny handful of prosecutors are responsible for a vastly disproportionate number of death sentences. The question that this disparity prompts is: Why?

This report analyzes the records of five of America’s deadliest head prosecutors.  Three of them personally obtained over 35 death sentences each: Joe Freeman Britt in North Carolina, Bob Macy in Oklahoma, and Donnie Myers in South Carolina.  These men shared an obsession with winning death sentences at almost any cost.  For example, Joe Freeman Britt, who committed misconduct in more than 36% of his death penalty prosecutions, said: “Within the breast of each of us burns a flame that constantly whispers in our ear ‘preserve life, preserve life, preserve life at any cost.’ It is the prosecutor’s job to extinguish that flame.” The remaining two prosecutors, Lynne Abraham (Philadelphia County, Pennsylvania) and Johnny Holmes (Harris County, Texas), did not personally prosecute as many death penalty cases as the three men above, but nonetheless oversaw the imposition of death sentences against a staggering 108 and 201 people, respectively, during their terms.

Of these five prosecutors, only one — Donnie Myers — remains in office, and he plans to retire at the end of the year. One of the most remarkable findings from our research is the fact that once these prosecutors and their protégés left their positions, death sentences dramatically declined in these jurisdictions — a pattern that has only become clear in the years since their departures.

We also highlight five additional prosecutors who came very close to becoming members of this notorious group.  These runners-up have egregious records in their own states, and like the prosecutors above, the striking drop in new death sentences that has occurred in their respective jurisdictions since their departures illustrates their outsized impact on the death penalty.

Unfortunately, the problem of personality-driven capital sentencing has continued beyond the tenure of these prosecutors.  Over the past fifteen years, prosecutors have pursued far fewer capital cases and juries have returned far fewer death sentences than in years past.  Indeed, in 2015, juries returned just 49 death sentences, the fewest in recent history.  This number represents an 84.4% drop from the 1996 high of 315 death verdicts.  However, in the increasingly small number of the counties that still actively sentence people to death, a handful of prosecutors dominate death-sentencing statistics.

In the final section of this report, we offer a snapshot of three active prosecutors who, if they continue on their current trajectories, may soon join the ranks of the deadliest prosecutors in America.  Taken together, the profiles featured in this report demonstrate that the death penalty has been, and continues to be, a personality-driven system with very few safeguards against misconduct and frequent abuse of power, a fact that seriously undermines its legitimacy.

June 30, 2016 in Data on sentencing, Death Penalty Reforms, Detailed sentencing data, Who Sentences? | Permalink | Comments (5)

Wednesday, June 29, 2016

US Sentencing Commission publishes "Overview of Federal Criminal Cases – Fiscal Year 2015"

On Monday, the US Sentencing Commission released this new data report, excitingly titled "Overview of Federal Criminal Cases – Fiscal Year 2015."   This USSC webpage provides this summary of the report's contents and findings:

The United States Sentencing Commission received information on 71,184 federal criminal cases in which the offender was sentenced in fiscal year 2015. Among these cases, 71,003 involved an individual offender and 181 involved a corporation or other “organizational” offender. The Commission also received information on 24,743 cases in which the court resentenced the offender or modified the sentence that had been previously imposed. This publication provides an overview of those cases [and includes these key findings]:

  • The 71,003 individual original cases reported to the Commission in fiscal year 2015 represent a decrease of 4,833 (6.4%) cases from fiscal year 2014.

  • Drug cases continued to be the most common type of federal case.  The 22,631 drug cases reported to the Commission in fiscal year 2015 accounted for 31.8 percent of all cases report to the Commission.

  • Immigration cases were the next most common, accounting for 29.3 percent of the total federal caseload.  In fiscal year 2011, immigration cases were the most common federal crime; however, since that year the number of these cases has steadily declined.

  • In fiscal year 2015, an imprisonment sentence was imposed on 87.3 percent of all offenders. Another 7.2 percent of offenders received a sentence of probation (i.e., where no type of confinement was imposed), a rate that has decreased over time from a high of 15.3 percent in 1990.

  • Almost three-quarters of offenders sentenced in fiscal year 2015 received a sentence of less than five years.

  • Methamphetamine offenses were the most common drug trafficking offenses and were the most severely punished drug crime in fiscal year 2015.

  • The proportion of drug offenders convicted of an offense carrying a mandatory minimum penalty was the lowest it has been since 1993.

June 29, 2016 in Data on sentencing, Detailed sentencing data, Drug Offense Sentencing, Federal Sentencing Guidelines, Mandatory minimum sentencing statutes, Offense Characteristics | Permalink | Comments (0)

Tuesday, June 28, 2016

"The Criminal Justice Black Box"

The title of this post is the title of this notable new paper authored by Samuel Wiseman and now available via SSRN. Here is the abstract:

“Big data” — the collection and statistical analysis of numerous digital data points — has transformed the commercial and policy realms, changing firms’ understanding of consumer behavior and improving problems ranging from traffic congestion to drug interactions.  In the criminal justice field, police now use data from widely-dispersed monitoring equipment, crime databases, and statistical analysis to predict where and when crimes will occur, and police body cameras have the potential to both provide key evidence and reduce misconduct.

But in many jurisdictions, digital access to basic criminal court records remains surprisingly limited, and, in contrast to the civil context, no lucrative market for the data (apart from that for background checks) exists to induce the private sector to step in to fill the gap.  As a result, bulk criminal justice data is largely limited to survey data collected by the Bureau of Justice Statistics.  Unlocking the “black box” by uniformly collecting and reporting basic, anonymized data from criminal cases — including, e.g., the charges, pretrial release decision, appointment of counsel, and case disposition — would have significant benefits. It would allow researchers, reformers, and government actors to both more effectively study the system as a whole and to more easily identify jurisdictions violating the Constitution by, for example, routinely denying counsel or pretrial release and imprisoning defendants for inability (rather than unwillingness) to pay a fine or fee.

This Article documents this problem, explores its causes, and proposes a solution, arguing that the federal government should form a framework for the uniform collection of anonymized local, state, and federal criminal justice data. While participation in this uniform system is likely to be incremental, even partial data would improve our understanding of the system as a whole and aid efforts to enforce well established, but frequently violated, constitutional rights.

June 28, 2016 in Data on sentencing, Detailed sentencing data, Who Sentences? | Permalink | Comments (0)

Tuesday, June 21, 2016

Bureau of Justice Statistics releases new detailed report on recidivism of federal offenders

This official press release reports on some of the interesting highlights of this interesting new report from the Bureau of Justice Statistics about recidivism rates and patterns for federal offenders.  The report is formally titled "Recidivism of Offenders Placed on Federal Community Supervision in 2005: Patterns from 2005 to 2010." Here is the text of the BJS press release on the report:

Of the nearly 43,000 federal offenders who were placed on federal community supervision in fiscal year 2005, an estimated 43 percent were arrested at least once within five years of their placement, the Bureau of Justice Statistics (BJS) announced today.  An estimated 18 percent of these offenders were arrested at least once within one year of placement on community supervision and 35 percent were arrested at least once within three years of placement.

An estimated 80 percent of offenders who were placed on federal community supervision in 2005 were male.  More than a third (41 percent) were white and nearly a third (31 percent) were black. An estimated 28 percent were age 29 or younger and about 42 percent were age 40 or older.

The first arrest offense for federal offenders after placement on community supervision varied by federal and nonfederal offenses.  Among federal offenses, public order offenses, such as probation violations, accounted for 90 percent of first arrests of federal offenders after placement on community supervision, compared to 33 percent of first arrests for nonfederal offenses.

In comparing federal and state prisoners placed on community supervision, almost half (47 percent) of federal prisoners were arrested within five years, compared to more than three-quarters (77 percent) of state prisoners. Nearly a third (32 percent) of federal prisoners returned to prison within five years of their release to community supervision, compared to more than half (59 percent) of the state prisoners.

Other findings include —

  • Nearly a quarter (23 percent) of federal offenders on community supervision were directly sentenced to probation, while more than three-quarters (77 percent) began a term of community supervision following release from prison.

  • An estimated 70 percent of federal offenders on community supervision had at least one prior nonfederal arrest, and more than a third (35 percent) had four or more prior nonfederal arrests.

June 21, 2016 in Data on sentencing, Detailed sentencing data, Offender Characteristics, Reentry and community supervision | Permalink | Comments (4)

Thursday, June 16, 2016

"States of Incarceration: The Global Context 2016"

The title of this post is the title of this notable new report from the folks at the Prison Policy Initiative. This press release from PPI provides an overview of the context and contents of this report:

How does your state compare to the international community when it comes to the use of incarceration? Not very well, says a new report and infographic by the Prison Policy Initiative.

“When compared against each other, some U.S. states appear to be far more restrained in their use of incarceration than high incarcerators like Louisiana,” said Peter Wagner, Executive Director of the Prison Policy Initiative and co-author of the report. “But all U.S. states are out of step with the rest of the world.”

This report, “States of Incarceration: The Global Context 2016,” updates our 2014 briefing that, for the first time, directly situated individual U.S. states in the global context.

“Massachusetts and Vermont have the lowest incarceration rates in the U.S.,” said Alison Walsh, report co-author and Policy & Communications Associate. “Compared to Louisiana, these states look progressive. But if these states were independent nations, they would rank as the 11th and 12th greatest users of incarceration on the planet, following the United States and a group of nations whose recent history often includes wars, military coups and genocides.”

The report includes an interactive graphic showing the incarceration rates for individual U.S. states and the District of Columbia and all countries with a population of at least 500,000. The report also includes a separate graphic comparing the incarceration rates of the U.S. to several NATO nations. “I hope that this data helps all states prioritize further criminal justice reforms. Lower incarceration rates are not only possible, in the rest of the world they are a reality,” said Wagner.

The report and infographic draw international figures on incarceration from the Institute for Criminal Policy Research’s World Prison Brief and state-level figures from the Bureau of Justice Statistics, the Bureau of Prisons and the U.S. Census Bureau.

The Easthampton, Massachusetts-based Prison Policy Initiative was founded in 2001 to expose the broader harm of mass criminalization and spark advocacy campaigns to create a more just society. The organization is most well known for sparking the movement to end prison gerrymandering and for its big picture data visualization “Mass Incarceration: The Whole Pie.”

June 16, 2016 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Scope of Imprisonment, Who Sentences? | Permalink | Comments (0)

Friday, June 10, 2016

"The Color of Justice: Racial and Ethnic Disparity in State Prisons"

The title of this post is the title of this notable data-heavy new report from The Sentencing Project.  Here is part of the reports "Overview" section:

Growing awareness of America’s failed experiment with mass incarceration has prompted changes at the state and federal level that aim to reduce the scale of imprisonment. Lawmakers and practitioners are proposing “smart on crime” approaches to public safety that favor alternatives to incarceration and reduce odds of recidivism.  As a result of strategic reforms across the criminal justice spectrum, combined with steadily declining crime rates since the mid-1990s, prison populations have begun to stabilize and even decline slightly after decades of unprecedented growth. In states such as New Jersey, New York, Rhode Island, and California, prison depopulation has been substantial, declining by 20-30%.  Still, America maintains its distinction as the world leader in its use of incarceration, including more than 1.3 million people held in state prisons around the country.

At the same time of productive bipartisan discussions about improving criminal justice policies and reducing prison populations, the U.S. continues to grapple with troubling racial tensions.  The focus of most recent concern lies in regular reports of police brutality against people of color, some of which have resulted in deaths of black men by law enforcement officers after little or no apparent provocation.

Truly meaningful reforms to the criminal justice system cannot be accomplished without acknowledgement of racial and ethnic disparities in the prison system, and focused attention on reduction of disparities. Since the majority of people in prison are sentenced at the state level rather than the federal level, it is critical to understand the variation in racial and ethnic composition across states, and the policies and the day-to-day practices that contribute to this variance.  Incarceration creates a host of collateral consequences that include restricted employment prospects, housing instability, family disruption, stigma, and disenfranchisement.  These consequences set individuals back by imposing new punishments after prison.  Collateral consequences are felt disproportionately by people of color, and because of concentrations of poverty and imprisonment in certain jurisdictions, it is now the case that entire communities experience these negative effects.  Evidence suggests that some individuals are incarcerated not solely because of their crime, but because of racially disparate policies, beliefs, and practices, rendering these collateral consequences all the more troubling.  An unwarranted level of incarceration that worsens racial disparities is problematic not only for the impacted group, but for society as whole, weakening the justice system’s potential and undermining perceptions of justice.

This report documents the rates of incarceration for whites, African Americans, and Hispanics, providing racial and ethnic composition as well as rates of disparity for each state.

June 10, 2016 in Data on sentencing, Detailed sentencing data, Prisons and prisoners, Procedure and Proof at Sentencing, Race, Class, and Gender, Scope of Imprisonment | Permalink | Comments (2)