« SCOTUS overturns "OJ death sentence" in Snyder | Main | Around the blogosphere »

March 19, 2008

Great new paper on prosecutorial choices in Mizzou

This new paper from SSRN provides fitting companion reading to today's Supreme Court work in Snyder (basics here).  The piece is entitled "Life and Death Decisions: Prosecutorial Discretion and Capital Punishment in Missouri," and here is the abstract:

This article presents the results of an empirical study of intentional homicide cases in Missouri. The authors created a database of 1046 cases; it includes substantially all of the homicide cases prosecuted in Missouri over a five year period that were initially charged as murder or voluntary manslaughter and that yielded criminal convictions. The authors selected 247 cases from the larger database for more detailed analysis.  We analyzed geographic and racial disparities in the rates at which: prosecutors charge first-degree murder versus lesser charges; prosecutors seek the death penalty, not lesser punishments; defendants are convicted of first-degree murder versus lesser crimes; and defendants are sentenced to death, not lesser punishments.

The Missouri statute gives prosecutors very broad discretion.  We estimate that at least 76 percent of the cases in the database are death-eligible under the statute.  However, prosecutors pursued capital trials in only about five percent of the cases.  Thus, death-eligible cases in which prosecutors chose not to pursue capital trials comprise at least 71 percent of the cases in the database.  Prosecutors in different counties exercise their discretion differently, leading to substantial variation in charging and sentencing practices in different counties across the state. The analysis of cases by race of victim and race of defendant shows that there are racial disparities in charging and sentencing decisions, but the racial disparities are not as significant as the geographic disparities.  The article presents measures of racial and geographic disparities without controlling for individual culpability; a follow-on study will introduce culpability measures as control variables.

March 19, 2008 at 05:08 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Great new paper on prosecutorial choices in Mizzou:


"Prosecutors in different counties exercise their discretion differently...."

That is technically called "democracy."

The reason why we elect prosecutors locally is to empower the people to influence how the criminal law is administered in their area.

Anyone who thinks that statewide uniformity is preferable should start a movement to abolish separate district attorney offices and have all prosecution done by a single, centrally commanded department of justice, as in the federal government.

Posted by: Kent Scheidegger | Mar 19, 2008 5:49:43 PM

Here's another notable data item from the paper you and others might find notable, Kent. From page 5:

"Overall, black defendants were less likely to be sentenced to death, although the sentencing rates depended significantly on the race of the victim as well, with defendants who kill white victims sentenced to death more often than those who kill black victims."

Posted by: Doug B. | Mar 19, 2008 7:55:32 PM

Doug, when one controls for geography, does the victim disparity go away? Likely it does.

Posted by: federalist | Mar 19, 2008 8:47:38 PM

That data item would be consistent with a decision by the people of the counties with higher black populations to elect prosecutors who seek the death penalty less often and decisions of the juries of those counties to impose it less often. That is what we saw in Maryland. The "race of the victim" effect dropped below statistical significance when controlled for jurisdiction. I haven't had time to review this paper yet, but that is what we have seen elsewhere: local democracy and jury of the vicinage working as designed, creating a "disparity" that disappears when legitimate variables are controlled.

Posted by: Kent Scheidegger | Mar 19, 2008 9:15:41 PM

Kent, I think one other issue too is the per murder resources of particular jurisdictions. A large city simply cannot seek death as often (on a per murder basis) as a suburb can. I don't think it's just a function of the ballot box and the jury box.

I will say this, it is the height of sophistry, to the extent the victim disparity is caused by "jury box/ballot box" sensitivity to the views of populations with higher percentages of minorities, to cite the race of the victim effect as "racist".

Posted by: federalist | Mar 19, 2008 9:42:13 PM

It is frequently (though not by any means always) the case that the murderer and the victim are of the same race.

So if death is sought more often in black on black and less in white on white, I imagine this would be described as racist because prosecutors are harder on black murderers.

But if death is sought more often in white on white and less in black on black, this can also be described as racist, as prosecutors are not putting sufficient value on the lives of innocent black victims.

Similar arguments can be made for the rate at which juries agree to impose death (racist jurors . . .) or the rate at which death sentences are overturned on appeal (racist judges), and so on.

The bar is being set awfully high.

Posted by: William Jockusch | Mar 20, 2008 1:51:04 AM

Mr. Jockusch -- Just curious: When the issue is the state intentionally taking the life of a person necessarily incapacitated by incarceration, who therefore presents no current threat to the public, how low should we drop the bar?

Posted by: Dean | Mar 21, 2008 12:39:28 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