Judge v. Jury
Forget what you know about who’s more likely to convict
Posted Jun 5, 2007 1:06 AM CST
By Jason Krause
Andy Leipold, a professor at the University of Illinois College of Law, was researching the right to a jury trial when a statistic involving federal criminal cases jumped out at him.
“I checked one year first and saw that judges acquitted more often than juries, but I figured it must be an anomaly,” he says. “Then I saw the same thing for every year I looked at--and realized this was something unexpected.”
The figures contradicted one of the oldest pieces of conventional wisdom in the legal profession: That defendants get more sympathy from a jury than a judge.
After some more digging, Leipold also discovered a correlation between judicial acquittal rates and the introduction of strict federal sentencing guidelines two decades ago. Could it be that judges, concerned that uniform required punishments were too severe for the specific crime, would rather set defendants free?
In 2005, Leipold published a research paper, “Why Are Federal Judges So Acquittal-Prone?” in the Washington University Law Quarterly. But he has since continued to follow the statistical trend in federal criminal trials.
Using data from the Administrative Office of the U.S. Courts, Leipold found that, since 1946, federal juries and federal judges have convicted defendants at almost the same rate: 75 percent by juries and 73 percent by judges.
But barely a year after the introduction of federal sentencing guidelines, judges and juries began heading in different directions. In the 14 years from 1989 through 2002, the conviction rate of federal juries increased to 84 percent, while that of federal judges decreased to 55 percent. In 2006, jury conviction rates exceeded bench rates by 25 percentage points (89 percent to 64 percent, respectively).
“It’s fascinating and it’s not consistent with my personal experience,” says Barry Boss, office managing partner of Cozen O’Connor’s Washington, D.C., office and co-chair of the ABA’s Criminal Justice Section Sentencing Committee. “The conventional wisdom has always been that jurors give more breaks or are more distrustful of authority or the system than judges, and are less likely to believe authority figures like police officers.”
While these findings may be counterintuitive --and irrelevant to state criminal courts--even Leipold says lawyers shouldn’t ditch the jury for a bench trial. The statistics should be just one set of factors to consider when plotting a litigation strategy in federal court. “I want to be modest in what I claim here,” Leipold says. “This is potentially significant information, but attorneys should use the information in making a decision rather than reflexively choosing a jury trial.”
There remains some question whether stiff federal sentencing guidelines have, in fact, influenced judges to become more lenient. But that question yields another, newer one: Now that the U.S. Supreme Court has made sentencing guidelines advisory in U.S. v. Booker, 543 U.S. 220 (2005), will the trend survive?
Reaction to reform?
There are reasons to believe that judges, at least in federal court, may in fact feel they have cause to give defendants a more sympathetic hearing. In particular, the Sentencing Reform Act of 1984 and the subsequent limitations on judicial sentencing discretion after 1987 may have made judges more temperate.
Though Leipold is cautious in his interpretation of the statistics he found, there seems to be a near-perfect correlation. Since the federal sentencing guidelines came into effect in the late 1980s, the trend toward higher rates of bench acquittals has only increased. “The drop-off is so dramatic and the timing coincides with the guidelines so well that it just hits you in the face,” says Leipold. “The timing is perfect and the judges’ dissatisfaction with the guidelines was so public that it’s hard to ignore.”
Before the Supreme Court intervened, a judge knew exactly what sentence a defendant faced, while juries were--and are--precluded from knowing. Defense attorneys often find ways to communicate this information at trial, but in most cases jurors decide on the evidence without knowing the punishment.
“It’s just human nature when there is something real significant at stake to be much more vigilant,” Boss says. “If you know the ramifications, you’re less likely to impose what you think may be a harsh sentence.”
Federal judges have openly complained that the guidelines put them in a virtual straitjacket. In 1996 a Federal Judicial Center survey found that 80 percent of the federal judges who responded were of the opinion they should be given more leeway in sentencing.
“It’s always possible that if judges think the sentence is disproportionate to the offense, they’re going to be more lenient,” says Frank Bullock, a retired U.S. district judge now in private practice with Womble Carlyle Sandridge & Rice in Winston-Salem, N.C. There are a number of alternative explanations for the shift. Judges may be hearing weaker cases for the prosecution, or only cases the defense is certain it will win. But if that is the case, that trend has not translated into more dismissals, as one might expect with more weak cases coming to court.
On the other hand, defense attorneys may be going to juries with weaker cases in the hopes of winning over only one or two jurors.
When Leipold separated out the types of crimes in question--violent, property, drug, regulatory and securities, immigration and public order (which includes tax, obstruction, and traffic offenses on federal property)--the correlation survived. Though the percentage of convictions varied from crime to crime, juries were very consistent, convicting more than 80 percent of defendants in all classifications except regulatory crimes (which was at 67 percent).
Although their conviction rates varied widely between categories, judges still offered lower conviction rates across the board. In public order crimes, which constituted nearly two-thirds of the bench trials reviewed, judges acquitted more often than not with a 47 percent conviction rate.
If public order crimes are eliminated, the 1989 to 2002 bench conviction rate rises from 51 percent to 60 percent. In particular, traffic cases--incidents on federal land of which there were more than 9,000 in federal court since 1989--figured heavily in this number. Leipold believes this figure helps explain the reason judges are more acquittal-prone. If you break out the public order crimes by seriousness, judges are again more lenient with misdemeanors, acquitting about half the time.
“I think if you consider these cases with minor charges, it would seem even more likely that a judge will react even more strongly against imposing a mandatory sentence,” Leipold says.
However, he also found that drug and weapons sentences, which have been considered unduly harsh among the judiciary, did not translate into a more pronounced gap in acquittal rates. “I would have guessed that in cases involving strict mandatory minimum sentences, the judicial conviction rate would show a big change,” says Leipold. “There was a difference, but not a blowout.”
It’s impossible to know whether judges are adjusting the standard of proof they require in a case depending on the severity of the punishment, but it is a reasonable inference, Leipold says. But rather than implying a reaction against harsh sentences, the statistics point to a more general dissatisfaction with the inflexibility of the guidelines.
And there were some genuine limitations to the study. In particular, Leipold was unable to break defendant statistics down by age, race, gender, criminal history, education or other information. Other factors to be considered are what type of representation a defendant had--whether it was a public defender or private attorney--and in what jurisdictions the cases were heard.
And of course, prosecutors weigh the same equations and have the power to deny bench trials in most systems. If they see defense attorneys turning to judges more often, it’s reasonable to think they’ll deny more such requests.
What About Booker?
Adding to the ambiguity of the data is the potential effect of the 2005 Booker ruling. That decision found key parts of the federal sentencing guidelines unconstitutional, in particular sentencing enhancements based on facts neither admitted by the defendant nor decided by a jury at trial.
It would be logical to assume that in a post-Booker world, with judges given more discretion in sentencing, conviction rates from the bench will likely rise. However, Booker is not likely to be the final word in sentencing as there are several related cases before the Supreme Court this term. And Congress has yet to re-enter the sentencing debate.
Others suggest that acquittal rates are probably not the most important statistics to consider. Karen Redmond, public information officer with the federal courts in Washington, D.C., points out that 90 percent of all federal cases end in conviction. She says that number is also influenced by the federal sentencing guidelines, whose mandatory sentences tend to scare more defendants into dealing for lesser charges rather than going to trial. The real trend, she says, has been away from the courtroom altogether.
“In total in 24 years on the bench, I had only one or two bench trials, so I just don’t think the information is very useful,” says Bullock, the former federal jurist. “Attorneys have different motives for waiving the jury trial, but I didn’t think it was a good idea to go to a bench trial. If you have a choice, I just have a better feeling about 12 people rather than one hearing a case.”
Knowing that federal judges may be more likely to acquit than juries should be considered when weighing a defendant’s options, but it is unlikely defense attorneys will be rushing to bench trials.
“Trial work is very much an art and not a science,” Boss says.
“Anytime new data is available, you have to factor it into your decision. But there are no general rules.”
Jason Krause is a legal affairs writer for the ABA Journal.