The Paradox of Punishment by Thomas J. Miceli

Author:Thomas J. Miceli
Language: eng
Format: epub
ISBN: 9783030316952
Publisher: Springer International Publishing

(5.5)

where ε0 is the highest defendant type willing to accept the deal. Now suppose the prosecutor offers a longer sentence in exchange for a plea, that is, she increases b. According to (5.5), this should raise the threshold of defendants who will accept the offer, that is, ε0 should rise as well. This will have two effects. First, the trial rate will increase because fewer defendants will accept plea deals. Second, the conviction rate at trial will also increase because some defendants who previously accepted a plea deal now opt for trial, and these marginal defendants necessarily have a higher probability of conviction than all those who previously opted for trial under the lower plea offer, thus causing the average conviction rate to rise.14 The result is a predicted positive correlation between trial rates and conviction rates. Conversely, if the prosecutor has superior information, there would be a predicted negative correlation between trial and conviction rates. Using data from U.S. District Courts from the early 1960s through the mid-1980s, Froeb (1993) found a positive correlation for most crimes, thereby providing support for the sorting outcome consistent with superior information on the part of defendants.

This conclusion raises the question of whether the use of plea bargaining as a way to sort defendants is desirable from a welfare perspective. This is an important question because sorting does not arise in this model as a result of a conscious effort by prosecutors, but as a by-product of asymmetric information, given the self-interested behavior of prosecutors and defendants. This brings us back to the crucial question concerning what motivates prosecutors. Suppose that prosecutors are in fact public-spirited and, in contrast to the above model, consciously seek to use plea bargaining as a sorting device as a way to reduce legal error. Would that be a desirable strategy?

Grossman and Katz (1983) have proposed such a theory for the scenario where defendants have private information about their chances at trial. In their model, they make the further simplifying assumption that the probability of conviction at trial for factually guilty defendants is uniformly higher than the probability of conviction for factually innocent defendants. In other words, they assumed that the two distributions in Fig. 5.1 are degenerate at their means, where recall εG > εI. In this case, a plea offer b can always be found such that

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.