Same Crime, Different Sentence

One of the many interesting questions in the world of post-Booker federal sentencing is whether a judge, in sentencing one defendant, is required (or permitted) to take into account the sentences imposed on codefendants.  Yesterday, the Third Circuit issued an interesting opinion on this question.  (Doug Berman has an excerpt and link to the opinion here.)  In United States v. Parker, the defendant received a sentence of 349 months, while his codefendants in the drug trafficking case received only 86 and 180 months.  Parker argued this his sentence was "unreasonable" (the post-Booker standard for appellate review of sentences) "because it failed to take into account 'the need to avoid unwarranted sentence disparities mong defendants with similar records who have been found guilty of similar conduct' as provided by [18 U.S.C.] 3553(a)(6)." 

I have argued elsewhere that (a)(6) should not be read to encompass codefendant disparities, so I was not unhappy that the Third Circuit rejected Parker's argument.  Yet, while the court did not require codefendant sentences to be considered, the court did indicate that judges were permitted to take the factor into account.  I am a big fan of increased sentencing discretion post-Booker, but I am troubled by the prospect of a sentencing judge saying something like this: "I see, Mr. Defendant, that your guidelines sentence is about 20 years, but your coconspirator, whom I sentenced a few months ago, got only 10 years.  So, I'll split the difference and give you 15."  (For an example of a real case in which the judge seemed to be doing just that (United States v. Strange), see my article linked above.)  This approach seems to make the sentence depend on the vagaries of who gets prosecuted and sentenced first--precisely the sort of arbitrariness that the guidelines were intended to eliminate.

Where I would find codefendant disparity the most palatable as a sentencing factor would be where the disparity highlights some systematic problem in the guidelines sentencing calculus.  For instance, Section 3E1.1 of the guidelines operates as a de facto trial penalty, meaning that the defendant who invokes his constitutional right to a trial is likely to get a longer guidelines sentence than his codefendant who pled out.  Such disparities are troubling, but what makes them troubling is not that two codefendants received different sentences per se, but that they received different sentences based on a consideration that shouldn't really matter.  To the extent that district courts take up the Third Circuit's invitation to consider codefendant disparities, I hope they will do so with careful attention to what actually drives the disparities, rather than taking the more clumsy approach of Strange.