The Volokh Conspiracy

[Andy Leipold (guest-blogging), June 26, 2006 at 10:13am] Trackbacks
Acquittal Prone Federal Judges

Many thanks to Eugene for inviting me to make some comments on some recent research.

A few years ago I noticed in the Sourcebook of Criminal Justice Statistics that federal defendants who stand trial are much more likely to be acquitted in a bench trial than by a jury. This seemed odd to me – I had always assumed the opposite was true. So I studied government records for federal trials between 1989 and 2002 and found a number of surprising things.

First, I found that the gap between bench acquittal rates and jury acquittal rates was quite large: over the 14 years I studied, the average conviction rate in jury cases was 84%, while judges convicted slightly more than 50% of the time. Second (using other data), I found that this gap was a recent phenomenon. Between the early 1960s and late 1980s, the conviction rates for judge and jury was roughly the same; the 20 years before that, judges actually convicted much more often than juries.

So the goal was to try to explain why this “acquittal gap” between bench and jury exists, and secondarily, why it had grown so large since the late 1980s. To make sure I wasn’t stumbling around in an academic fog, I started by interviewing two dozen federal prosecutors and defense counsel to see if their instincts were the same as mine. They were: of the 24 lawyers I spoke to, only a very few knew or guessed that judges are more likely to acquit. The rest were mildly to strongly certain that juries were more favorable to the accused.

During the coming week discuss some of the variables I looked at -- type of case, seriousness of charge, type of defense lawyer, strength of the evidence, etc.), as well as some of the conclusions I drew. Questions, comments, and criticisms are welcome, aleipold@law.uiuc.edu. The full paper can be found at 83 Wash.U.L.Q. 151, from SSRN at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=843606.

anonyomousss (mail):
here's a question from a reader: why start by characterizing the issue as about acquittal-prone judges rather than conviction-prone juries? or using a neutral phrasing like "difference in acquittal rates?" your characterization smacks of biased research.
6.26.2006 10:17am
Hans Bader (mail):
Maybe this is because jurors are more likely to acquit the guilty (who comprise most criminal defendants), while experienced judges are more likely to acquit those who are actually innocent (who comprise a minority of criminal defendants). Experienced judges are better at sifting the guilty from the innocent.

Thus, the innocent may seek bench trials, while the guilty seek jury trials. And there may be fairly few bench trials, because most criminal defendants are guilty, and their attorneys would rather have their cases heard by jurors, who tend to both acquit more guilty jurors even though they convict more innocent ones.

Frankly, outside of a few areas, such as drug conspiracy cases (where false accusations to get a plea bargain are common), certain types of fraud cases, and date-rape and child-molestation cases, there aren't many innocent criminal defendants.

So when you ask a criminal defense attorney who is more likely to acquit their client, they're thinking of who is most likely to acquit the average (i.e., guilty) defendant, not an innocent defendant.
6.26.2006 10:30am
byomtov (mail):
My guess matches Hans'. I think there may be a strong selection mechanism at work. If jury verdicts are seen as less accurate than those of judges this is exactly the outcome we would expect.
6.26.2006 10:37am
18 USC 1030 (mail):
My initial thought would be that the innocent would request a bench trial-- especially in cases in which a rather complex matter of law is at hand. One would, if innocent, rather have a judge interpret statute than a jury. In many cases, however, jury trials are more issues of fact rather than of law. That is, at least in my over-generalization, many defenses at jury trials are "I didn't do it." Whereas at bench trial it would be "I did it, but it wasn't illegal."
6.26.2006 10:38am
Sasha (mail):
Anonymousss -- I interpreted the title as ironic, since juries are the ones with the reputation of being "lawless."
6.26.2006 10:44am
KRM (mail) (www):
Good thought, but that still wouldn't explain why the shift from the 60's to the 90's, unless the cases have grown in complexity since. That would difficult to measure...
6.26.2006 10:47am
John M. Perkins (mail):
Until recently, [2005] there were almost no public defenders in Middle Georgia. Before that, only Houston County had such. Houston County is the home of Warner Robins.
6.26.2006 10:56am
AppSocRes (mail):
Based partly on the time period when this phenomenon emerged, I am guessing that type of case is going to be an important explanatory variable. I'll bet a lot of the variation is going to be found in drug cases. Judges seem very unhappy with the harsh sentences they must hand out in these cases and how frequently the evidence is primarily plea-bargain deals cut with co-conspirators. It wouldn't surprise me to find that judges have reacted with more frequent acquittals in these types of cases.
6.26.2006 11:05am
Steve Lubet (mail):
Hans Bader said: "Maybe this is because jurors are more likely to acquit the guilty (who comprise most criminal defendants), while experienced judges are more likely to acquit those who are actually innocent (who comprise a minority of criminal defendants)."

What is your basis for that observation? What makes you think that experienced judges are better at separating the guilty from the innocent? Isn't it possible that judges become jaded or cynical over the years?

You might be right, but I don't know how you can be so certain. If we had a reliable method of determining which defendants are "actually innocent," wouldn't we use that instead of holding trials?
6.26.2006 11:07am
LizardBreath (mail):
Hans and byomtov have it, I think. While it might not be all innocent defendants, there's going to be a class of innocent defendants who have defenses that are very strong, but which may be difficult for a jury of laymen to follow. If a defendant in that situation has a good lawyer, they're going to end up in a bench trial and probably with an acquittal.

The argument isn't that judges are necessarily more or less accurate, but that they're more predictible, and so a defendant with a case that will appeal to a judge more than to a jury is likely to know this and be able to choose appropriately.
6.26.2006 11:21am
Hattio (mail):
My guess would back up the thoughts of 18 USC 1030. Fact questions probably go to a jury more often, and legal questions go to judges. Another thought, how are the acquittals being measured, per charge or as a whole? If per charge, judges may be better at recognizing over-charging by the prosecutor, and so reasonably only convict on 2/3 of the charges.
6.26.2006 11:25am
eeyn524:
What makes you think that experienced judges are better at separating the guilty from the innocent?

That case that was linked here a few weeks ago, where the jury decided a case by flipping a coin. I'd say that jury at least didn't understand the difference between "proved beyond reasonable doubt" and "what the hell maybe"; one would think most judges have better idea.
6.26.2006 11:50am
Malvolio:
of the 24 lawyers I spoke to, only a very few knew or guessed that judges are more likely to acquit.
Wow, that's scary. I mean, the basic point of a lawyer is to win for his side, and most of them are ignorant of a factor that more than triples the acquittal rate? Eek!
6.26.2006 12:35pm
JerryW (mail):
"Frankly, outside of a few areas, such as drug conspiracy cases (where false accusations to get a plea bargain are common), certain types of fraud cases, and date-rape and child-molestation cases, there aren't many innocent criminal defendants."



My medical practice is one town from Newark NJ. I have many city employed defense attorneys as patients. To make life more interesting I usually ask patients questions about their professional lives. I have uniformly asked what percent of their personal clients are guilty. The number 90% is amazingly consistent.
6.26.2006 1:01pm
John Herbison (mail):
I would suspect that the strength of the government's case has a lot to do with the willingness of both sides to waive a jury. In those jurisdictions that require unanymity to convict, each juror may practice de facto nullification. In a case where the evidence is strong, that inures to the benefit of the accused, who would accordingly be less likely to waive a jury.

Marginal cases are winnowed out in the charging process or, in many cases, are resolved by some kind of plea agreement. I suspect that bench trials most often occur where the facts are not in dispute and the question is one of the proper application of law to the facts.
6.26.2006 1:14pm
Steve:
Wow, that's scary. I mean, the basic point of a lawyer is to win for his side, and most of them are ignorant of a factor that more than triples the acquittal rate? Eek!

This observation ignores the strong argument made by Hans, above, and assumes that correlation equates to causation. It may well be that innocent defendants are more likely to prefer a bench trial and that guilty defendants prefer the confusion attendant to a jury trial. It does not follow that a given defendant has three times the chance of acquittal if he opts for a bench trial.
6.26.2006 1:16pm
Jason Fliegel (mail):
Based partly on the time period when this phenomenon emerged, I am guessing that type of case is going to be an important explanatory variable. I'll bet a lot of the variation is going to be found in drug cases. Judges seem very unhappy with the harsh sentences they must hand out in these cases and how frequently the evidence is primarily plea-bargain deals cut with co-conspirators. It wouldn't surprise me to find that judges have reacted with more frequent acquittals in these types of cases.


That was my first thought as well. One way to test this would be to compare the percentages that Professor Leipold found with corresponding percentages in state courts.
6.26.2006 1:33pm
john (mail):
Its not at all scary that most lawyers wouldn't know the difference between all jury trials and bench trials. Such a choice is made for tactical reasons on the basis of the facts in the case and the nature of the defendant. What all judges do in the aggragate is not overly relevant to what the judge in the current case will do.

Now why is there the big difference in conviction rates? From what I have seen, juries are more likley to put someone away if they look dangerous than are judges (any one judge perhaps not, but as a group anyway). Also I think HOW lawyers decide on whether to go to a jury or have a bench trial has a significant impact in the outcome. And it might be that because lawyers think the judges will be more likely to convict, some cases that would do better as a bench trial are shifted to a jury trial. Also having an impact is that since lawywers think judges are more likely to convict, perhaps lawyers only choose bench trials when they have a high level of confidence of a win (meaning that judges convict less because they only get cases where a conviction is less likely).
6.26.2006 1:47pm
jallgor (mail):
On a somewhat related note I vaguely recall reading a study in torts class where they gave a certain number of judges and juries identical written fact patterns and asked them to presume laibility on the part of the defendant. They then asked them to set a damages amount and the judges' awards were consistently higher than the jury awards. Obviously, I don't know or recall enough to judge the validity of the study but, like the conviction rates discussed in the above post, I think the result is exactly contrary to most people's expectations and perhaps related.
6.26.2006 2:36pm
Matthew Dundon (mail):
Don't forget the truth that everyone in the criminal justice system knows: crime makes you stupid. A lot of jury trials are driven by a defendant's irrational rejection of a reasonable plea bargain in favor of the street "wisdom" that "all it takes is one juror." A lawyer has to go to the jury trial if that's what his client demands, no matter if the client has a grossly mistaken understanding of who is likely to be empaneled and how effectively cop-skeptics are marginalized in deliberations and ultimately find themselves voting to convicct.
6.26.2006 2:40pm
KevinM:
I think it's a self-selection error. In the federal system, the prosecution never requests a bench trial, but may consent if the defendant requests it. But defendants will request a bench trial in cases where counsel believe strongly that the client will get a better shake from the judge. Often it's because of community resentment for the defendant or the particular kind of offense; other times it's because the defense is technical or counterintuitive. Defendant cannot unilaterally waive a jury; the consent of the prosecution and the approval of the court are required. Fed. R. Crim. P. 23(a). Nevertheless, at least when I was prosecuting, consent was almost invariably given. Fairness to the defendant, deference to the court, respect for the court's workload, and other factors all weighed in favor of consent.
Net result: bench trials only when defendant really wanted it. You'd expect a higher rate of acquittal in those cases.
6.26.2006 3:03pm
Eric James Stone (mail) (www):
Back when I was in law school in the early 1990s, one of my professors said that if your client is guilty, you should have a jury trial; if he's innocent, you should have a bench trial. The rationale was that juries were less predictable.

That fits in with what several commenters have mentioned above.
6.26.2006 4:01pm
Harry Eagar (mail):
I'm with Malvolio. Regardless of the explanation for the difference, if it's that big -- 8:5 -- and has changed that much over a fairly short period, you'd think both prosecutors and defenders would have noticed and would attempt to take advantage of it.
6.26.2006 4:07pm
Le Messurier (mail):
My own empirical observation is that this is not confined to the Federal Courts at all. Here in Maricopa, County, Arizona (Phoenix) it is rare to see a jury aquit, while the remainder of the cases are tried before a judge result in plea bargins as well as lighter sentences (or so it would appear.) 90 percent of all criminal cases in this jurisdiction are plea bargined. If this results in lighter sentences then this would go a long way to explaining why accused choose a bench trial by accepting a plea bargin. For instance, if a 3 count indictment is plea bargined down to 1 count with a lower class of felony which results in a lighter sentence, then why would one choose a trial by jury? The risk is too great; and all 3 counts would be tried by jury. Also, jury trials are expensive. To reduce this expense I'm sure there is tendency to "push" a quicker and cheaper solution, i.e the bench trial. The way to push the bench trial is to offer plea bargins that are too good to refuse.
6.26.2006 5:40pm
bellisaurius (mail):
I've gotten through to about page 172 when I felt the need to post some thoughts:

1.It looks like the current balance of bench vs jury is based on the last big study done in 1945. People saw the reverse numbers and sort of responded at the time. We will probably see another shift after this one since people are natural game theorists. You lawyer types are probably very intuitive about that topic.

2. (Page 166) Looking at the actual trend in the numbers of bench trail vs conviction rate, and the jury trial vs conviction rate, I was surprised to see the rose or fell in step with each other. As the number of bench trails fell, so did the conviction rate. # of jury trials rose, and the conviction rate rose. Obviously, this shows cases moving en masse.

3.(pp 171 and 172, the most interesting part)

Before I talk about this, I recommend looking at this article entitled: "why is third and two a passing down?"
http://www.pro-football-reference.com/blog/wordpress/?p=50
It covers the concept that these two pages show.

The conviction rate of crimes across the board (property, violent, regulatory, etc..) are almost the same in jury trials, about 80% -except regulatory, at about 60%- Bench trials are lower across the board, the closest is a couple percent difference in the regulatory category. Page 172 shows what's going on though. People do move trial to categories that they believe have better rates, in this case, the trial that people believe the bench will be lenient isn't the regulatory one, but the Public Order category with roughly 45% of those being bench trials (the rest are a little below 20%). In regulatory (the type I would think most laymen wouldn't understand), 82% go to jury, which is the average split between jury and bench.

I still have more reading, but pp171 and 172 are intersting in themselves.
6.26.2006 8:42pm