Is The Trial Penalty A Myth?

In a very provocative claim, Penn lawprof David Abrams contends that we’ve been wrong this whole time. Via Dan Markel at PrawfsBlawg:

Almost every teacher of criminal procedure is aware of the idea of the “trial penalty,” which conveys the sense that defendants who exercise their right to a trial will invariably get a worse result if convicted than if they plea bargain.

Comes now (or relatively recently at least) David Abrams from Penn with an article that slays the sacred cow of the trial penalty by providing, you know, data. And the data is the best kind of data because inasmuch as it’s true, it is SURPRISING data. Specifically, Abrams argues that based on the study he performed (which originally appeared in JELS and now appears in a more accessible form in Duquesne Law Review), the data supports the view that in fact there’s a trial discount not a trial penalty.

Whoa. Before you rush to court to announce ready for trial, read Abrams’ piece.  It’s short, but nonetheless headache inducing for lawyers who didn’t really enjoy statistics in college.  To call the foundation for Abrams’ argument data is to stretch its meaning a bit too far.  It’s a statistical approach, and empirical in that respect, but fails in the way that almost every attempt to compare apples to Fords fails.

Why care?

Abrams offers some suggestions for what might explain this surprise: possibly a salience/availability bias on the part of the lawyers who remember the long penalties imposed after dramatic trials. Regardless of what explains the conventional wisdom, the competing claims should be ventilated in virtually every crim pro adjudication course.

In other words, Markel urges lawprofs to teach their students that the “conventional wisdom” of the trial penalty may not be true, and there may actually be a trial discount instead.  Because law students already leave school with such a refined grasp of practice in the real world.

The universe of Abrams’ analysis, coming from Cook County, Illinois, includes dismissed cases, which he treats the same as acquittals after trial.  Given the pyramid of prosecutions, with the bottom being overloaded with crap arrests of no merit, where defendants are arrested just to teach them the cost of the ride, if not the rap, and dismissed at arraignment or shortly thereafter, his data is so horribly skewed as to render the balance of his analysis meaningless.

In addition, Abrams neglects to account for pleas of convenience, where a defendant is held without the ability to make bail, facing a lengthy period of pre-trial custody followed by the uncertain outcome of a trial, and chooses to cop a plea that gets him out sooner rather than later regardless of guilt.

The issue of the trial penalty can’t be considered in the sanitized vacuum of statistical analysis, unless the real world influences that skew the results are taken into account, whether by eliminating them from the mix or somehow adjusting for the false result.  The far better way to approach an effort to determine the existence of, and extent of, the trial tax (or discount) is to compare head to head, the plea offer made versus the sentence after trial.

Of course, given how few cases go to trial anymore, this would produce too small a universe to draw much of a conclusion.  In addition, the study would then have to take into account additional factors, such as the nature of the crime, the quality of the evidence and the effectiveness of the representation.  It’s not that it’s easy, or even possible to do so, but that if you want to claim that conventional wisdom is wrong, then you need to have the goods to do so.

While criminal defense lawyers’ experience is easy to dismiss as anecdotal, and suffers from the same flaws as all anecdotal evidence, it’s not as if there is any criminal defense lawyer anywhere who has ever said, “why no, there’s no trial tax in my experience.”  That makes for a whole lot of anecdotal evidence, and indeed is the experience of every criminal defense lawyer everywhere.

It’s not that the rare case results in a sentence after trial below the plea offer. It’s happened to me, and no doubt has happened to many others.  But this is the outlier.  We go to trial when a case is winnable, meaning that we have some reasonable expectation that we can defend against the charges.  When a defendant is dead in the water, the equation changes provided the plea offer isn’t unduly onerous.

But when faced with evidence where the defendant has little reasonable chance of prevailing and a plea offer below the mandatory minimum sentence available for a conviction, it’s hard to comprehend how this doesn’t constitute a trial tax. Years of a person’s life are real, not subject to statistical regression analysis.  And no statistics will make the evidence go away.

Had Abrams attempted his study by limiting it only to indicted defendants, he might have latched onto something. Rid the study of the bottom of the pyramid, as they were never really in jeopardy to begin with, and focus instead on those whose cases are real and who are ultimately forced to make the decision whether to go to trial or take a plea.

In the meantime, this study is wholly unpersuasive, because of its failure to take into account the real world forces that drive the system, as well as the overwhelming anecdotal experience of every criminal defense lawyer ever.  It’s nice to fall back on data and offer an empirical analysis that says everything we’ve ever believed is wrong, but this isn’t it.

On the other hand, there is a bone in my head that makes me want Abrams to be right, so that we can tell defendants to go to trial.  But my duty to counsel clients truthfully precludes me from doing what would be better for me than for him.

16 thoughts on “Is The Trial Penalty A Myth?

  1. Jeff

    The author clearly has little to no actual practical experience in the criminal defense world — or prosecution for that matter. Just about every step in the analysis in the paper is problematic. My favorite is at the very end when he asserts the issue can be solved by rotating court staff.

    1. SHG Post author

      I don’t suspect that he sees practical experience as having anything to do given the nature of his analysis. If anything, he probably sees his perspective as unbiased by experience. But what comes through clear as a bell is that his analysis hasn’t accounted for the aspects that would make it real or illusory. Thus, he comes up with irrelevant numbers and silly solutions.

  2. william doriss

    Bravo, bravo!
    The trial tax is alive and well, as any CDL or defendant who has stood in the well of a full-blown court-trial can attest to. Or should be able to attest to!?! My CT cases are typical and instrumental/instructive, even if I say so myself. (Previously posted multiple times, here and there.)
    No, I did not read Abram’s treatise. Nor should I have to; nor am I inclined to. It is simpley out of the question, it’s absurd. Statistics lie, and liars can figure, as the cliche goes,…
    Hey look, assistant state’s attorney’s are typically ill, “mentally ill”–pardon my French. It comes with the territory. Why would anyone who is not a sadistic, narcissistic, hysterical oligarchical/patriarchical/autocratical/didactical patrician-dictator wannabe even consider this occupation? (When there are so many honest ways of making a living in the good ol’ U.S. of A.)
    John Kerry’s first job out of law school was Asst. State’s Attorney for Middlesex Co., Mass. That says it all in my book.
    P.S., I met John Kerry–who took MY place at Yale, for real–and he would not give me the time of day, nor a copy of the Constitution, or the consideration of a phone call. I rest. Check out my new photo on Twitter. You will not be disappointed. Grrrr. Where’s Barleycorn when I Kneed him?

  3. Charlesmorrison

    To answer the question/title of the post: No. I didn’t read the original piece, as I don’t need a new headache. But to suggest such a fact is utterly ridiculous; I don’t care what the numbers say. And, as you noted in the post -dismissals are not acquittals. How in the world could you conflate or combine the two? By definition, one does not involve a trial and is irrelevant to evaluating punishment after trial.

    While I appreciate that social science is difficult, quantitatively, the most pertinent data is not always capable of capturing: what the judge says he’ll accept as a sentence in chambers for a plea versus what he actually imposes after a jury convicts.

    While you hit upon the need to utilize much more relevant data, not every plea offer (agreed range, agreed cap or agreed determinate sentence) is placed on the record in every jurisdiction, notwithstanding the Frye decision. Some are “informal” discussions as to how to resolve the case.

    Your right to suggest that every case is different… Hell, the facts at trial can actually help in some circumstances, but the problem with this claim, as you mention, is that it lacks the only relevant data (offer versus actual sentence). But one aspect not addressed is that much of the relevant data is off the record and can’t be captured no matter what amount of digging someone is willing to do.

    And, just because the state agrees that X years is acceptable on the record, doesn’t cure the problem that judges can decide to not accept an offer. Who hasn’t had a deal rejected? Are researchers going to search to see whether a judge ultimately said “no way”?

      1. Charlesmorrison

        Almost always the case. But, this is akin to saying the world is flat in a headline. I don’t need to read that.

        1. SHG Post author

          I appreciate your point, and would have felt the same way. I read it because it truly confused me how it was possible that he had data to refute a point that was so clear to me. Having done so, I see where he was coming from, realize how it could be that he was so wrong and yet appreciate what he was trying to do. It’s still wrong, but I got it. It was worth thinking about things differently, even if I wasn’t persuaded.

  4. John Barleycorn

    Game Theory, Odyssey Chickens, and Clothed Prosecutors. An Expected Value Analysis.

    It would have been nice to see a quarter page of footnotes addressing the EV probabilities of stacked charges crossing the road.

    His 10% math and “seriousness of the charges” need a little more attention. I was also disappointed he did not attempt to quantify every prosecutors unspoken desire to be loved.

    But the Odyssey Chicken reference alone was worth the read.

  5. Pingback: Police and prosecution roundup - Overlawyered

  6. Mark Draughn

    You say Abrams treats dismissals the same as acquittals after trial. That would be devastating to his conclusions — a dismissal is more like an ultimate plea bargaining collapse than a trial result — but I don’t see where you get that from his paper. Am I missing something in the legal language?

  7. Marc R

    Oh what great timing! I’m picking a jury next week where the judge previously inquired if I was sure my client didn’t want to enter a guilty plea and avoid trial. Whereas I previously worried he was hinting at a trial tax, I now realize my past experience is mere poppycock balderdash and will advise my client immediately of this new groundbreaking study. Scott, you’re letting your 50 years of trial experience cloud your ability to see mathematical proof. Sorry, I’ll take math over warm fuzzy feelings of some amorphous “punishment.”

Comments are closed.