« Triple killer in North Carolina taunts he is "gentleman of leisure" of death row | Main | Ex-HealthSouth Chief Scrushy gets a year off prior sentence at resentencing »

January 25, 2012

Intriguing story of federal sentencing guidelines producing disparity

One of many reasons I do not think it always sound or ever truly wise to use non-guideline sentencing decisions as a metric of increases federal sentencing disparity is because there are ways in which the federa; guidelines can sometimes produce disparity.  An interesting example of this reality comes from this recent press report from Alabama, which is headlined "Two Alabama bank robbers, 2 different sentences: Judge laments 'quirk' in federal guidelines."  Here are the details:

Two bank robbers with similar profiles appeared before U.S. District Judge Kristi DuBose in Mobile last week, but due to what the judge termed a "quirk" in the sentencing guidelines, one walked out with a lighter sentence.

Prior criminal convictions earned Jason Paul Davis, 26, the title of "career criminal," under the law, and the DuBose sentenced him to 12 years and 7 months in prison for holding up banks in Selma and Tennessee.

She noted that the other man, William Normer Overstreet, actually has a longer criminal record than Davis.  Like Davis, he threatened a teller during his heist.  Like Davis, he led authorities on a high-speed chase.   And like Davis, he blamed his crime on a longtime drug addiction.

But one of the 48-year-old defendant’s convictions was too old to count against him, so he was not considered a "career criminal" under the guidelines.  The recommended punishment was much less.  "You need to convince me not to go with the government’s recommendation for the highest end of the guidelines," DuBose told defense attorney Larry Moorer, who was in the middle of making his case for leniency.

Citing the punishment she imposed on Davis the previous day, DuBose publicly toyed with the idea of sentencing Overstreet to a prison term longer than the guideline range.  In the end, she settled on a prison term at the top of that range — 9 years and 7 months.

January 25, 2012 at 11:08 AM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Intriguing story of federal sentencing guidelines producing disparity:


If one of his convictions was considered "ancient" and was properly not counted toward a career criminal designation(which carries a higher guideline range) how has a disparity occured? Clearly, the two offenders were not similarly situated.

The court had the discretion to sentence the second offender outside the guideline range if his prior record warranted same.

Posted by: mjs | Jan 25, 2012 11:20:42 AM

Fair point, mjs, though it sounds as though they were are similar as two bank robbers with lengthy criminal histories are likely to be. Of course, you are right the judge could have gone outside the guidelines to create parity here, but that's my point. If she had, then the second sentence would have been recorded as a non-guideline sentence and there would be a data point suggesting "disparity" simply based on one less within-guideline sentence. Put another way, I am always concerned about lies, damn lies, and guidelines statistics.

Posted by: Doug B. | Jan 25, 2012 12:43:06 PM

The judge sure could have gave him the 12 yr 7 mon sentence....I think what she did was tough, but fair.. She gave him the high end of the range, that made that old charge still have some weight...Armed robbery, is not to be taken lightly and this judge did her fair share..

Posted by: Josh2 | Jan 25, 2012 3:04:13 PM

This type of disparity arises more frequently in cases where prior convictions are treated differently under the categorical approach. For instance, Defendant A has two prior convictions for simple possession of large amounts of cocaine, but he doesn't qualify as a career offender because the convictions are for simple possession. Defendant B, on the other hand, has two prior convictions for possession with intent to deliver tiny amounts of marijuana, so Defendant B is a career offender, because those are drug trafficking offenses.Similar difficulties arise regarding the categorization of crimes of violence.

It's kind of hard to drum up a lot of sympathy for a bank robber who engages in violent behavior, but the starkness of these disparities really shows up in drug cases. A rather insignificant federal drug trafficking case can result in a decade-long sentence for Defendant B in the above scenario, while the same offense might result in a rather insignificant sentence for Defendant A.

Of course, this is a well-known problem under the guidelines. The preferable solution depends on whose ox is being gored.

Posted by: C.A.J. | Jan 25, 2012 3:35:13 PM


Really nice example of how the guidelines' reliance on so many technical concepts with regard to criminal history produce disparity. Past charging and sentencing decisions carry forward into the criminal history calculation. Then you have the previously discussed (in an earlier posting here) issue of instant charging decisions leading to disparity. The irony is that the guidelines are masquerading as a system of parity, but in fact are structured to result in frequent disparity, some of which is the result of deliberate manipulation of its provisions (most often by the Government). The only solution, in my humble opinion, is to try to get more Judges and stubborn intellectually lazy USPOs (yep, I'm ragging on many of my colleagues!) to fully embrace the simple state of the law: These guidelines are advisory. They are no more important than the other 3553(a) factors. They often miss the mark as to individualized sentencing.

As to the bank robbers here, if the Commission had decided to count prior sentences going back 20 or 25 years, the would both have been Career Offenders. If the Commission had decided to count back only 5 years, neither of them would have been. I'd love to see the empirical basis for the time lines they chose. I'd also love to see a Dodo Bird make sweet love to a Griffin....

Posted by: USPO | Jan 25, 2012 4:48:44 PM

Doug: You are wise to be skeptical of statistics but I don't know of any claim that sentences outside the guidelines are evidence of disparity.

In my humble opinion, disparity occurs when similarly-situated offenders receive different sentences. Cases taken outside the guidelines theoretically fall outside the "heartland" of cases and are distinguishable from the customary set of facts.

If the number of non-guideline sentences becomes excessive, the Commission would look to modify the guideline to bring it inline with prevailing judicial practice.

Posted by: mjs | Jan 25, 2012 4:55:22 PM

DOJ and many others often cites to different rates of non-guideline sentences in different districts when expressing concerns about increased disparity after Booker. The point of this post is to encourage a skeptical view of any such claims (which I sure will continue to get made).

Posted by: Doug B. | Jan 25, 2012 5:15:38 PM

USPO - Thanks for telling it like it is. You are exactly right.

Posted by: AFPD | Jan 26, 2012 12:25:21 AM

C.A.J., I actually don't think your example is a good one. If someone has a "tiny" amount of marijuana, it is usually extremely difficult to prove intent to distribute. The defendant will have a persuasive argument that it was a personal use amount, absent additional facts. The opposite is usually true for "large amounts" of cocaine. I'm not sure what you consider a large amount, but let's go with a kilo. Kinda hard to say that's for personal use. Intent to distribute isn't hard to prove with an amount like that.

Posted by: domino | Jan 26, 2012 7:41:20 AM


You mischaracterize "the simple state of the law" to suggest that the guidelines are no more important other 3553(a) factors. An accurate calculation of the guidelines is the starting point for all federal sentencing.

Posted by: mjs | Jan 26, 2012 11:19:09 AM


Actually, it is quite common to run across convictions for possession with intent to distribute, or sale, of small amounts of marijuana as well as other drugs. Usually these are the results of undercover purchases. When someone actually sells drugs, no matter how small the amount, to an undercover officer or a confidential informant, it's not hard to show distribution.

It is just as common to run across convictions for simple possession of surprisingly large amounts of all types of drugs. A defendant can receive a lengthy sentence for simple possession of a large amount of drugs, so the thinking is presumably, why bother charging an extra intent element when we can so easily show possession? Also, some states--I believe North Carolina is one--punishes people as drug traffickers based on the amount of drugs possessed, even if intent to distribute is not shown. In federal court, these cases would still be treated under the guidelines as simple possession, and they would not contribute to a career offender designation under the guidelines.

Understandably, some judges are troubled by this approach. They don't like it that they can't just presume a possession conviction is a drug trafficking conviction if it involves, say, multiple kilos of marijuana. But I believe the approach is based on the principle that we don't want to have to go relitigate old cases. And it works both ways--there are, in fact, real world cases of people being convicted of possession with intent to distribute truly minuscule amounts of drugs. Those defendants would appreciate an opportunity to show a federal court that they weren't really selling drugs, notwithstanding the conviction. But the categorical approach also prevents relitigating their cases.

Posted by: C.A.J. | Jan 26, 2012 12:02:55 PM

I'm not sure why it's 15 years. If you did it 14 years ago, that's only incrementally less time, 7.5 percent less, than someone who did a crime 15 years ago. It would make more sense to me to have it be a shorter period of time, but reset the count when someone committed a crime again within that period. And start it at the release date. From the previous sentence. I suspect that the 15 years was arrived at by considering the average criminal justice sentence in many cases. I don't know. The whole system seems unwieldy.

Posted by: Bill B. | Jan 26, 2012 5:49:05 PM


I totally agree that the Court must properly calculate the advisory range before engaging in any analysis of the appropriate sentence. But after that...
I don't know what Circuit you are in, but here in the 9th, the District Court cannot presume that a sentence within the advisory range is reasonable, or that a sentence outside the range is unreasonable. Further, the Court of Appeals cannot begin with any such presumption either (although I understand in some Circuits, they can, but not the District Court). Also, the plain construction of 3553(a) does not afford the advisory range any kind of "super factor" status. Frankly, I am tired of the advisory range as starting point being conflated with the notion that it is the presumptively appropriate ending point. Especially in light of Kimbrough and its progeny. What's your basis for disagreeing?

Posted by: USPO | Jan 26, 2012 8:58:09 PM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB