A new study finds that although wrongfully convicted may theoretically receive either higher or lower punishments compared to correctly convicted defendants, the evidence suggests that the average punishment of wrongfully convicted defendants is lower
EDIT: this post was updated on 19.12.18. A link to the working paper can be found below.
Wrongful convictions are gut-wrenching. It is therefore not surprising that recent year pop-culture has taken a special interest in true-crime stories of defendants who claimed that they have been wrongfully convicted (see, for example, Netflix's "The Staircase" and "Making a Murderer" and the podcast "Serial").
Punishing the innocent
While many are concerned about the prospect of a wrongful conviction, less attention has been given to the punishment of wrongfully convicted defendants. There is of course a straightforward reason why this is the case: we do not know who is actually wrongfully convicted. If we did, the wrongful conviction would probably never have occurred to begin with.
Furthermore, we typically assume that the punishment should fit the crime - such that two defendants convicted of the same punishment should receive the same punishment.
The Law and Economics of Punishments
However, the story is a bit more complex than that, because of how sentencing works. In fact, there is reason to suspect that the innocents may receive either a higher or a lower punishment. For example, innocents are less likely to take a plea bargain (which typically includes a reduced sentence) and are less likely to express remorse during sentencing. Thus, innocents may get higher punishments on average.
On the other hand, judges might suspect that a defendant is innocent, but nonetheless believe that the evidence is sufficient to convict. Then, judges may choose to convict but give a lower sentence - such that if they were wrong, the injustice 'isn't that bad.' Thus, innocents may get lower punishments on average.
In my new paper, which is now available on SSRN, I develop a theoretical model which encompasses these arguments and shows how wrongfully convicted defendants may receive different punishments and how this feeds back into crime deterrence. In a nutshell, I identify four effects that impact the judges' decision during sentencing and conviction: cost of wrongful convictions, utility from correct convictions, utility from facing more guilty (and less innocent) defendants, and cost of wrongful acquittals. I then proceed to test the predictions of my model empirically.
How do can anyone test this?!
My empirical strategy involves contrasting the punishments of exonerated defendants (at the time of their conviction) with the punishment of other convicted defendants.
What are the results?
I find that wrongfully convicted defendants receive lower punishments on average, but only after controlling for plea bargains. That is, the unconditional punishment handed out to wrongfully convicted defendants is actually higher, but that is only because wrongfully convicted defendants - as a group - are less likely to take a plea bargain.