The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent


Detrimental Reliance and Stare Decisis

Insightful thoughts from Dean Vik Amar relevant to Ramos v. Louisiana


One factor the Court frequently recites in deciding whether to overturn a previous precedent is "reliance." In what way have people relied on the perhaps-incorrect precedent that might counsel against overturning it? But there is a lot of confusion over what kind of reliance interests count and exactly how to think about them.

One place this came up recently was in the Court's decision in Ramos v. Louisiana, ending the practice of allowing non-unanimous juries to convict defendants of crimes—and again in the Court's refusal to make that decision retroactive in Edwards v. Vannoy. In both cases, reliance interests might seem to be at stake, since the decision imperiled decades of practice used to obtain convictions in multiple states.

Dean Vik Amar wrote a recent column making a very insightful point, that what should matter to the analysis is not just reliance, but detrimental reliance:

[T]he reliance that must be protected via stare decisis—and that is protected in common law fields — is not reliance that leaves you surprised or disappointed, but reliance that leaves you worse off than you would have been had the earlier event—in this setting the mistaken earlier ruling — never occurred. That is what is meant by "detrimental," a word that is key to understanding most aspects of legal reliance that is unfortunately missing from much of the Court's stare decisis analyses.

For instance, part of the reason we protect reliance in property and contract is that a party might have made subsequent investments that are in turn premised on a previous understanding of the law. Similarly, overruling the Supreme Court's legal tender cases would frustrate detrimental reliance because people would lose not just the present and future rights to use paper currency in a particular way, but potentially tons of commercial investments already made.

And here Amar applies the analysis to Ramos and Edwards:

And applying this notion to the unanimous-jury-verdict cases, we see that Justice Kagan gets it wrong. What were the reliance interests at issue in Ramos? How could the state(s) argue that they would be worse off under a unanimous-jury-verdict rule today (if that be the correct constitutional interpretation of the Sixth Amendment and incorporation, points I do not take up today) than if Apodaca had never been decided? Certainly the reliance cannot take the form of the notion that states will be able to convict fewer people under a unanimity regime than under a rule permitting non-unanimity, or that convictions are more costly to obtain if they require unanimity. Those are costs to be sure, but they are the costs the (ostensibly correct) constitutional rule requires. Instead, the states can argue that they are worse off than they would have been absent Apodaca only insofar as—with respect to persons on direct appeal, the only ones at issue in Ramos—they could have tried and convicted those several hundred people who were convicted by non-unanimous verdicts under a unanimity rule the first time around, they will have to expend more resources by trying those folks twice rather than once (and perhaps some of those who would have been unanimously convicted had that been the rule in place the first time around will not be convicted because witnesses will have died or memories will have faded.) Non-trivial reliance, but not a huge deal either. So if jury unanimity in state verdicts is required by the Sixth and Fourteenth Amendments (again, questions I don't engage but that weren't the basis of Justice Kagan's stare decisis reasoning), then there was no reason to retain Apodaca.

What about this month's ruling in Edwards? Would applying Ramos retroactively harm states that reasonably and detrimentally relied on Apodaca? (And while the question whether to apply a new ruling retroactively is not technically the same doctrinal question as whether to overrule a past ruling, the two queries are very analytically similar, because they both ask how much of a past, mistaken ruling should continue to govern today, in light of detrimental reliance on it.) There is no doubt that applying Ramos retroactively would impose high costs on states that detrimentally relied on Apodaca. Having to retry tens if not hundreds of thousands of people whom a state might have convicted the first time around under a unanimity rule had such a rule been in place is tremendously expensive, to say nothing of the many defendants who would have been convicted unanimously years ago had that been the governing rule but who might not be convicted today because of lost evidence, dead witnesses, and stale memories. And yet here—where detrimental reliance by states is significant—Justice Kagan invokes stare decisis (that is, to give broad effect to the Ramos rule now on the books) to meaningfully hurt states for their reasonable reliance on Apodaca. This doesn't make sense if detrimental reliance is one of the keys to understanding stare decisis.

As Amar notes, we can imagine specific kinds of reliance on a criminal procedure decision like Ramos that might count as detrimental reliance—maybe the state convicted a defendant 11-1, but had they known about the Ramos rule the judge would have charged the jury to keep deliberating, and they would have eventually come around to a 12-0 conviction. Or maybe the defendant would have scored an 11-1 mistrial, but had the state known about the Ramos rule it would have immediately pressed for a second conviction while evidence and memories were fresh. But the dissents don't try to sketch out these specific arguments for detrimental reliance and trying to do so shows just how speculative they are.

I think it's no secret that there will be a lot more debate about overruling precedent in the next few years of the Roberts Court. This strikes me as a very useful clarification that ought to be part of the conversation.