« ICYMI: Ten (okay, Nineteen) Tips for New Law Professors | Main | Bad TV presentation of law (a weekly series?) »

Saturday, September 08, 2018

Elected Prosecutors and Non-Prosecution Policies

Rachel Rollins is likely to be the next district attorney in Suffolk County, Massachusetts.  She recently won the democratic primary, and she is not facing a Republican challenger in the general election.  Rollins is a reform-minded candidate, and so her win has caused a bit of a stir.  Rollins is also causing a stir because she recently released a list of crimes that she says she will decline to prosecute if elected.  The list includes larceny under $250, receiving stolen property, disorderly conduct, and other charges.

If you had asked me ten years ago what I thought about a prosecutor making a public announcement about all of the crimes she would not prosecute, I would have been extremely critical.  In particular, I would have said that prosecutors have a duty is to enforce the laws, and policies of non-enforcement conflict with that duty.  Just as decisions about what to criminalize belong to legislatures, I would have said, so too do decriminalization decisions belong to legislatures, not prosecutors.  But having spent time studying the relationship between criminal justice institutions, I am much less critical of Rollins’ announcement.  In particular, I no longer think that decriminalization decisions are reserved for the legislature.

To be clear, my point about decriminalization is a descriptive one, not necessarily a normative one.  Legislatures have—to a significant extent—delegated the scope of criminal law to prosecutors.  Legislatures have done so by writing overly broad criminal laws.  Legislatures write overly broad laws knowing that prosecutors will not prosecute all behavior that falls within the statutory text.  Prosecutors will make their own decisions about the circumstances under which to prosecute crimes.  In other words, prosecutors are free to decide what conduct to treat as legal and what to treat as permissible.  We got a glimpse of that decision-making when James Comey explained to Congress that, even if Hillary Clinton’s behavior fell within the text of a federal criminal statute, it did not meet the criteria that DOJ uses for actually bringing charges in such cases.

The sheer number of crimes also delegates questions about the scope of criminal law to prosecutors.  Resource constraints make it impossible for prosecutors to bring charges against everyone who commits a crime.  As a result, district attorneys have to make decisions about which crimes to prioritize and which crimes to deprioritize.

Some might respond that there is a difference between making enforcement decisions in the context of individual cases and making enforcement decisions ex ante about an entire category of cases.  But I don’t see the appeal of that argument.  The elected district attorney is the individual who has been empowered to make these important decisions.  So one hopes that she is making those decisions in the context of a general office-wide policy, rather than leaving the decision to line prosecutors in individual cases.  Line prosecutors may have different priorities, and case-by-case decisions would doubtlessly lead to inconsistent decisions.

Some might also object that, even if an elected prosecutor has formal policies about what charges not to bring, she should not announce those policies to the public.  If I know, for example, that my district attorney is not going to prosecute larceny of goods under $250, then (so the argument goes) I will feel free to steal items that cost less.  (Meir Dan-Cohen fans will recognize that argument as one of acoustic separation.)

Even assuming that such public announcements decrease deterrence—and I am uncertain whether they do—public announcements have a significant benefits  :  For example, public announcements can help ensure that law enforcement won’t enforce these laws in an arbitrary or discriminatory fashion.  One major problem with overly broad laws or rarely enforced laws is that they can be used as a pretext for racial discrimination or other animus.  In this respect, overly broad laws and rarely enforced laws are similar to vague laws

Most importantly, public announcements make prosecutors democratically accountable for their enforcement policies.  Unannounced policies do not allow the public to make informed voting decisions or to give feedback to elected prosecutors about how they have chosen to enforce the criminal laws.  But in announcing this list of crimes, Rollins has ensured some measure of legitimacy and accountability for her decisions. 

I am sure that at least some people will read this post and think that it has failed to fully justify or defend a practice of categorical non-enforcement.  There are very intelligent people who have argued against categorical non-enforcement.   In an ideal world, I would probably agree that legislatures should make all criminalization and decriminalization decisions.  In such a world, legislatures would write narrowly targeted criminal laws and would not criminalize more behavior than the system could reasonably prosecute and punish.  But that is not the world we live in.  Since legislatures have decided to delegate significant questions about the scope of criminal law to prosecutors, prosecutors should answer those questions in a transparent and accountable fashion.  That is what Rachel Rollins has done here.

Posted by Carissa Byrne Hessick on September 8, 2018 at 09:37 AM in Carissa Byrne Hessick, Criminal Law, Law and Politics | Permalink


Bill Otis writes, "Does the DA or the legislature have a better claim on the authority to make that decision?"

In the present system, the prosecutor. I agree that in an ideal, theoretical world the legislature should. But we don't live in that world. We live in a world where, as the trope goes, everyone commits three felonies a day. And that "three felony a day" world is a world that has been created by the legislature.

The problem here is of the legislature own making. It can solve this problem anytime it wants; it can repeal laws. Imagine there was only one law on the books...that would end any discretion the prosecution has about choosing which laws to enforce. One law, either enforce it or not.

Posted by: James | Sep 10, 2018 1:52:57 PM

More on Rachael Rollins indeed and the reforms proposed , here and the links therein :



Posted by: El roam | Sep 10, 2018 11:29:33 AM

More good comments. Three additional responses.

First, my defense of what Rollins has done here is a structural one--that is to say, I am defending the idea of making a list public, not the particular crimes that she included on her list. In fact, as I said in a previous comment, I am even more supportive of prosecutors releasing these lists when they will include crimes that I think should be prioritized. The release of the list allows a public discussion of the prosecutor's priorities. The failure to release a list does not.

Second, I appreciate Bill Otis's point about the executive duty to take care that the laws are faithfully executed. Zach Price's article on this topic (which I link to in the last paragraph of the post) makes an argument against these sorts of public statements on precisely those grounds. Zach's article is incredibly well-researched and well-written, and anyone who is interested in this topic should read it.

Zach argues that case-by-case non-enforcement is permissible, but general non-enforcement policies are not. Although Zach makes a very convincing historical case for the position that general non-enforcement policies run afoul of the Take Care Clause, his argument is significantly undercut by the reality of modern substantive criminal law. Substantive law criminalizes so much behavior, and there are insufficient resources to prosecute all crimes. Zach acknowledges this and he concedes that prosecutors have to adopt policies about when they will not enforce the laws. But Zach insists that those policies have to be confined to setting priorities, rather than making policy decisions about the wisdom or desirability of particular criminal laws--a distinction that he concedes is "more a matter of mindset than any sort of bright-line rule." In other words, Zach doesn't disagree that prosecutors should have non-enforcement policies, but he thinks that those policies have to arise from particular motives.

I admire Zach's attempt to make a nuanced and sophisticated argument about a complicated topic. But I think his concession shows a serious flaw with his argument. Prosecutors are obviously going to take policy matters (such as the wisdom or desirability of certain criminal laws) into account when setting their priorities. And Zach seems to approve of precisely that sort of decisionmaking when he grudgingly approves of the Obama-era policy of marijuana non-enforcement in states that had legalized pot.

So that is all a very long way of saying that, although I appreciate the appeal of the "take care argument," I don't think that it can be defended given the current state of our substantive criminal law.

Finally, let me address the deterrence argument that Bill Otis makes here, and that has gotten some folks on Twitter especially upset. I'm not an expert in what makes some people commit crimes and what makes others desist. But I do think that the available evidence cuts against the simple, rational actor deterrence model that is quite popular in public discourse. There is, for example, a lot of non-criminal behavior that I don't engage in: I don't engage in it because there are social sanctions associated with it, because it might open me up to civil liability, and because I have more attractive alternatives to that behavior. Put differently, the criminal law is not the only force at work in society. So let's please not pretend as though making behavior criminal is the only way to discourage people from engaging in that behavior.

What is more, I think that model is deeply problematic because it ignores the trade-offs associated with enforcement policies. If a prosecutor chooses not to enforce crime X, she will have more resources to devote to crime Y. Even if a public non-enforcement policy might increase the incidence of crime X, more enforcement of crime Y would decrease the incidence of that crime. If crime Y is more harmful than crime X, then we should applaud that decision. In any event, any increase in crime X should be judged in relation to any decrease of crime Y.

This comment has become long enough that it probably warranted its own post. But I really appreciate the opportunity to engage with everyone about this.

Posted by: CBHessick | Sep 10, 2018 9:41:32 AM

Carissa ,

As assumed by me earlier , It doesn't make sense simply , that the supreme court , has granted , absolute discretion whatsoever , to any prosecutor . In that case I brought , there was a delay of 18 month in the prosecution . The district court , and the circuit court , affirmed the prejudice , and dismissed indeed the indictment . The supreme court reversed , but it was based upon question of law or error of law primarily , that is to say , that according to the supreme court , delay of such ( before indictment ) doesn't violate the constitutional right , for speedy trial , and the due process . But, one could conclude that in other potential or theoretical cases , surly dismissal would be warranted , here I quote the opinion ( one dissenting by the way ) :

We are to determine only whether the action complained of - here, compelling respondent to stand trial after the Government delayed indictment to investigate further - violates those "fundamental conceptions of justice which lie at the base of our civil and political institutions," Mooney v. Holohan, 294 U.S. 103, 112 (1935), and which define "the community's sense of fair play and decency...

End of quotation :

So surly , if one defendant or suspect , would prove malice prima facie for example , then , surly courts would grant him standing , and intervene , can't be absolute . For the dissenting judge by the way , it was sufficient , that lacking any pretext or reasonable explanation in the records ( for such delay ) for dismissing the indictment .


Posted by: El roam | Sep 9, 2018 6:16:45 PM

Just he who wants , the syllabus to the case mentioned by me down there :



Posted by: El roam | Sep 9, 2018 5:20:00 PM

Prof. Hessick,

Let's test this a bit. Would you have the same approval if the DA announced that she would not prosecute larceny under $500? Under $2500? That she would not prosecute cocaine offenses? Statutory rape where the victim was 15 years old or older? Simple assault where the combatants were drunk and no weapon was used? A weapon was used but it wasn't a gun or a knife? That she would not prosecute malum prohitium offenses? Is there any articulable limit (other than what you can get by with in a one-party jurisdiction) on where this is headed?

Doesn't her policy invite small-time larceny, theft (since it will be easier for a thief to find a fence for the stolen goods), and public bullying? If I wanted to make a living being a small time thief, would I not be well-advised to move to Suffolk County?

It might make a degree of sense to adopt these things as low prosecution priorities, given a limited budget. But is it wise to announce them publicly? And in blanket form? And is it really about budgets, or is it about a substantive disagreement with the legislature about what ought to be made criminal? Does the DA or the legislature have a better claim on the authority to make that decision?

The DA is an officer of the Executive Branch. Does Massachusetts have a constitutional provision similar the that of the federal government, requiring the executive to take care that the laws be faithfully executed? If so, how does the future DA's decision square with that?

Finally, I hope to have the chance to meet you at one of the FedSoc events both of us frequently attend.


Bill Otis

Posted by: Bill Otis | Sep 9, 2018 5:18:52 PM

Meanwhile , One may read here a case in the supreme court ( 1977 ) touching it :



Posted by: El roam | Sep 9, 2018 5:14:55 PM

"Line prosecutors may have different priorities, and case-by-case decisions would doubtlessly lead to inconsistent decisions."

The problem with letting line prosecutors make these decisions isn't that it leads to inconsistent decisions but rather that it turns the formation of rules/standards over to unelected officials rather than being made by elected officials who can be held accountable in the democratic process. It is letting the tail wag the dog.

Posted by: James | Sep 9, 2018 4:44:34 PM

Carissa ,

I don't really think , that this is what the supreme court has decided . Doesn't make sens . Probably , it is more plausible , to argue , that it is indeed the prerogative of the prosecution , but if , prima facie , it is proven , that something is wrong with the prosecution ( I mean , on the face of it , for example , clearly malicious looks ) surly a defendant or suspect would have standing and judicial review . Don't think it is an absolute discretion . I shall check later more details . Meanwhile , you may read one here , titled :

Questioning the Autonomy of Prosecutorial
Charging Decisions: Recognizing the Need to
Exercise Discretion - Knowing There Will Be
Consequences for Crossing the Line , here :



Posted by: El roam | Sep 9, 2018 4:34:32 PM

Thanks for the interesting comments to this post. A couple of quick responses:

First, the idea of judicial review of decisions not to prosecute that El roam mentioned caught my eye. The Supreme Court has made clear that decisions not to prosecute are not subject to judicial review. But there are plenty of critics of that decision. James Vorenberg's 1981 article "Decent Restraint of Prosecutorial Power" is an especially good example of such criticism.

Second, in response to Orin's question, I'd say that the post is meant to highlight the weakness with the "don't usurp the legislative role" argument, rather than an endorsement of Rollins' particular list. Prosecutors ought to have policies about what charges they pursue, and under what circumstances--and my understanding is that many prosecutor offices do. But those lists ought to be public because the public ought to have input into what that list looks like.

I have little doubt that I disagree with the charging and declination policies of many prosecutors offices. But I would fully support those offices making those policies public. In fact, I am more invested on having the policies with which I disagree become public so that there can be a discussion surrounding those policies.

Thanks for the responses, everyone!

Posted by: CBHessick | Sep 9, 2018 4:03:50 PM

Carissa, provocative post. Just to clarify, though, are you saying you suppprt this announcement, or only that you are less critical of it than you would have been 10 years ago? Also, to what extent does your view hinge on the specific list of crimes that the candidate claims will not be enforced?

Posted by: Orin Kerr | Sep 9, 2018 10:35:52 AM

Actually, looking at what Miss Rollins has announced is that her office will have as default to decline prosecution of certain offenses, and that in some cases of non-prosecution she is open to treating them as civil infractions. That is rather different from the impression I got first, which was that she had released a list of crimes she would never prosecute. (The way fornication laws went completely unenforced long before they were held unconstitutional.)

The actual content of her policy does not seem terribly unusual in principle, though the actual content of her list may differ from what would be popular in other parts of the country.

Some of the crimes on her list I would frankly prefer were never prosecuted at all.

Posted by: Jr | Sep 8, 2018 6:14:36 PM

If prosecutors were not allowed prosecutorial discretion but had to bring charges whenever they could, I am pretty sure some of these overly broad criminal laws would be clarified and narrowed pretty soon. Whether that would be worth it is debatable of course.

Posted by: Jr | Sep 8, 2018 5:49:09 PM

Very interesting and important issue . Some may like it , some others not . But one thing should be beyond any debate :

And it is , the " judicial review " of course . Each case or decision of such , and the policy itself , must be subjected to potential judicial review . For , the post has missed the issue of victims . There are always almost , victims . They must have their say . They must have chance to challenge it in courts , or other tribunal , external or not associated with the department itself implementing the policy .

In many states in the world , it is a routine simply . Constitutional or administrative tribunal , hearing challenges to such decisions or policy , not to prosecute . In fact , many times , it can be challenged by " outsiders " not victims even ( in the US , typically , one must suffer personal injury in federal standing ) . That is to say , private entity , associations or alike , may challenge , in the name of the public , or public interest , such policy or concrete decision .


Posted by: El roam | Sep 8, 2018 11:04:05 AM

The comments to this entry are closed.