Federal Sentencing 2004

The United States Sentencing Commission statistics containing the 2004 numbers (impacted greatly by the demise of the binding nature of the Commission's federal sentencing guidelines) are out.

In the first half of 2004, 4.5% of criminal convictions arose from trials, the rest were plea bargained. This fell to a 3.4% trial rate in the second half of the year. The only offenses for which 10% or more of people convicted were convicted on the basis of trials, in either the early or later parts of 2004 (with the percentage of crimes in that category) were:

Murder 0.1%, Manslaughter 0.1%, Kidnapping 0.1%, Sex Abuse 0.4%, Arson 0.1%, Auto Theft 0.1%, National Defense 0.1%, Bribery 0.2%, Rackettering/Extortion 0.9%, Civil Rights 0.1%, Environmenta/Wildlife 0.2% (percentages do not add due to rounding errors).

The types of cases affected by the Sentencing Guidelines break down as follows (in order of frequency even when rounding obscures the differences in numbers, followed by the average number of months of the sentence in the later part of the year):

Drug Trafficing 33.7% (73.6)
Immigration 22.5% (21.8)
Firearms 11.6% (69.5)
Fraud 10.4% (13.6)
Larceny 3.1% (7.5)

Subtotal: 85.3%

Robbery 1.9% (93.6)
Drug Possession 1.9% (4.1)
Forgery/Counterfeiting 1.7% (12.4)
Administration of Justice 1.5% (14.5)
Pornography/Prostitution 1.3% (73.0)
Money Laundering 1.2% (29.9)
Embezzelment 0.9% (5.3)
Assault 0.9% (27.1)
Rackettering/Extortion 0.9% (65.0)
Tax 0.7% (12.4)
Prison Offenses 0.5% (16.1)
Drugs - Communications Facilities 0.5% (38.3)
Sex Abuse 0.4% (81.9)
Bribery 0.2% (11.7)
Environmental/Wildlife 0.2% (2.4)
Gambling/Lottery 0.1% (5.4)
Auto Theft 0.1% (28.6)

As is clear from these statistics, the people convicted in the federal system are overwhelmingly non-violent offenders.

When mandatory sentencing guidelines were in effect, 27.1% of people were sentenced to less than the guideline term (mostly at the request of the prosecution), while only 0.8% were sentenced to more than the guideline sentence.