Sunday, July 5, 2015

Two policy truisms

The faithful at Emanuel African Methodist Episcopal Church in Charleston, SC have reacted to the evil attack on their community with a quiet and loving strength and dignity. The remarks by the church’s Senior Bishop, Right Rev. John Richard Bryant, at the funeral service for the nine victims were particularly moving. And the families’ expressions of forgiveness to the shooter, Roof, were simply astonishing.

Contrast their response with what has been much more common in the media (on both sides of the political spectrum). I wish it weren’t so common to use tragedies like this one to advance a political agenda. It’s unseemly and smacks of opportunism. That said, I admit I’m hopeful that propriety won’t stop the developing head of steam to remove the Confederate flag from in front of the South Carolina Statehouse. Whatever else that flag is supposed to represent, it’s also a big, government sponsored middle finger to something like 28% of the state’s population.

A more important objection to folks who use tragedies like this to advance a political agenda is based on a presumption that political responses to relatively rare albeit tragic events will just make things worse. Various factors work together to get the policy through the legislative process despite plausible concerns about unforeseen and unintended bad consequences.

First, in the wake of such an event, natural emotional reactions – sympathy for the victims and fear related to being one – are running high. Second, this fear is all out of proportion to the probability of the event recurring. We’re just not that good at risk analysis. Limited, anecdotal evidence tends to weigh heavily in our evaluations of risk. And modern news reporting makes us think that certain kinds of events happen with a greater regularity than they really do. Third, elected politicians are subject to this same psychological pressure, but also a great deal of political pressure to respond. Would you want to be the one asking questions about the expected costs and benefits of, e.g., the Patriot Act after 9-11 or Caylee’s Law, let alone the one voting against them? Would you if you were contemplating an upcoming reelection campaign? Constituents want their representatives to DO SOMETHING, and a vote against the law is viewed as a vote against freedom or children or whatever. Sometimes bad, not-very-well-thought-out policy is good politics.

So here’s policy truism #1:
 Laws named after crime victims and dead people are usually a bad idea.
Many have used the Emanuel AME tragedy to argue for stricter gun control. I’m agnostic about the effectiveness of gun control generally and, specifically, about whether proposed changes to existing gun control laws would have prevented Roof’s murders. But I’m fairly certain that racial minorities will disproportionately suffer under a stricter gun control regime.

Last year, 48.6% of people convicted under federal firearms legislation were black. The racial disparity in this category of federal offense is greater than any other. Most proposed changes to firearms legislation would crack down further on illegal possession and sale of firearms, which comes with the need for police, agencies and prosecutors to exercise a great deal of discretion in how to go after these activities. Given the extent of violations, they can’t go after everyone. And we probably shouldn’t expect them to enforce new laws in an unbiased way when they already don’t enforce existing laws in an unbiased way.

New firearm offenses also mean new reasons to stop and search people that police tend to suspect of criminal activity or who present easy targets. In the end, there would be more people in prison, and that tends to mean more black men in prison. So the question isn’t whether it's broadly possible to have effective gun control without more and worse civil rights violations. The question is whether that's what we'd actually get.

Here then is policy truism #2:
The fact that a policy, if expertly tailored and benevolently administered, would make things better isn’t a good reason to implement it.

Kyle Swan
Department of Philosophy
Sacramento State


  1. Kyle, thanks very much for this interesting piece. I would add to what you say in the fourth paragraph that it is important for us to understand that in an increasingly transparent society where fewer and fewer misdeeds go unnoticed, it is a dead certainty that we will end up with an exaggerated conception of their frequency if we get our information mostly from the news.

    Given that most statutes that criminalize a behavior will disproportionately negatively affect the least well off in society (assuming the behavior is within their reach, so, not for example, laws against insider trading) I wonder how strong a consideration you think this is, in general. Do you think it is a strong consideration in favor of, say, repealing vagrancy laws? ( you count the regressive nature of taxes that discourage consumption (energy taxes, e.g.) as a strong reason not to implement them?

    The way I am reading you is this: Since new laws are almost certain to have unintended negative consequences (including uneven enforcement), we should exercise extreme caution in implementing them. I find that hard to argue with, but I think the realities of the relation between politics and legislation make dispassionate implementation a very tall order, and usually not a very realistic one. I am not sure what the basis of your agnosticism is with respect to the effectiveness of well-crafted gun control laws (especially after living for so long in Singapore!), but if you were not agnostic and you believed a law was well crafted to achieve its aims, would the fact that it is named to inspire approval and pushed in the wake of an information cascade be a good reason to oppose it?

    1. Thanks, Randy. I think regressive taxes are unfair, but the argument here is more based on the idea that there are likely to be problems when enforcement of the laws are discretionary in some way, and this is very likely to be the case when the violation of the law doesn't produce a victim, a complainant. Laws against possessing and selling drugs are other good examples (and it's no surprise that second greatest racial disparity in convictions of federal offenses is in drug offenses).

      The other focus of the argument is on the unforeseen/unintended consequences of the laws. When you have the relevant psychological factors and political incentives working in concert, the laws tend not to be well-crafted. But if it turns out they are (and if they are enforced fairly, as well) then, no, I don't think the provenance of the law by itself is sufficient to oppose it.

      By the way, I wouldn't sign off on Singapore's gun control laws. I don't think, for example, that illegal possession of a firearm should be punishable by caning or that illegal possession of more than one firearm should be presumed to be trafficking (which will get you hung).

    2. Kyle, thanks for that. Yes, I guessed when you said that you meant that you were agnostic about the effectiveness of gun control laws you meant something more complicated than that, but I couldn't resist.

  2. Kyle,
    thanks for this thoughtful reflection on the need for thoughtful reflection before legislating. I wonder, though, if your premises in support of Truism #1 and #2 actually wouldn't make it nearly impossible in a country like this to craft restrictive laws of almost any sort, though. My worry is not that we have too much darned liberty, but that a whole lot of really bad people doing really horrible things could argue similarly, "Well, you shouldn't outlaw X, not because I do X for a living, but that so many black men will be disproportionately blamed for doing X. It's really in their interest that X not be prohibited." When really everyone would be much better off if X were stopped or at least made really really hard to engage in. We ought to focus our concern for the poor, marginalized, despised, persecuted, and maligned, by demanding the police stop enforcing laws in a racist manner and holding them accountable for failing to do so. I mean, why should that even have to be said as a question of justice, let alone be used as an argument against otherwise reasonable legislation?

    1. Thanks, Chris. I worried about both those points. But I don't think the argument raises concerns for laws of any sort. I suggested that it raises concerns for certain kinds of laws (ones where it's difficult to identify a particular victim or a complainant; ones, for example, that criminalize selling and possessing things).

      Obviously we must do what you say we must do. But the answer to your question is that a good argument for a regulation isn't made simply by showing it would be a good idea. It's only made if you can also show that it will be implemented in a way that won't make things worse.

  3. Kyle, that seems like the right answer to Chris' question, and yet intuitively there is something unsettling about lumping, say laws against drugs and prostitution, which arguably makes things worse by forcing the behavior underground, and gun regulation which you suggest will make things worse because of uneven enforcement by police and the courts (not, of course, to suggest that concerns about the underground market aren't just as strong for firearms, they just aren't a focal point in your post.)

    When we think about the fact that a large number of people will continue to use illegal drugs, we have the further thought that, hey, well, maybe that's because people have a perfect right to do that sort of thing to themselves if they choose and we should think hard before messing with it. But we don't have any such intuitions about uneven enforcement, that's just messed up.

    I think what I'm suggesting is that if we make these into two clearly distinct categories- harms resulting from people arguably exercising their right to personal liberty and harms resulting from unfair discrimination, the intuitive pull of your argument is lessened to some degree. That is we may feel more inclined to go with gun regulation and simultaneously commit to rectifying the police abuse problem than go with drug regulation and commit to fixing the drug abuse problem.

    1. I don't think I see how that works. The basis for lumping the laws together is the fact that, whatever else is true, they all make things worse because of the way they're enforced. Before criminalizing something, of course we should think about whether people have a right to exercise their liberty in that way. But we should also think about whether enforcing the criminal sanctions will increase various abuses, including racial disparities in the criminal justice system. I think either of these thoughts is sufficient to worry about drug and prostitution laws. And I think the second thought is sufficient to worry about gun laws. Why wouldn't it be?

    2. My thinking is just that by using the general category "makes things worse," it makes it easy to draw specious analogies between kinds of legislation we disapprove of. I disapprove of some drug and vice laws because they make things worse, so now you tell me, yeah, see, that's just like gun control, it makes things worse, too. But I object to drug and vice laws because they make things worse in a particular way, one that I regard as more or less unavoidable in a liberal society. I do not object to them because corrupt or inept enforcement causes them to be administered in a discriminatory way. On that score, I object to the bad enforcement, and I believe that a liberal society can and should be committed to its eradication.

      Someone with exactly your views could have written a post arguing that if we are going to have increased gun control we need to be aware of the fact that black men are going to be on the receiving end of it and this will make it all the more important to continue to push for greater transparency in police practices, etc.

    3. Someone with exactly my views did write such a post. It was me. Except I did the tollens thing since pushing for greater transparency hasn't worked and isn't likely to work for public choicey reasons.

      Anyway, I understood the part of your thinking you described in your first paragraph. But the analogy isn't specious because besides the reason for disapproving of drug and vice law because people should generally be at liberty to do those things, you should also disapprove of them because the war on drugs and vice has led to a militarized force that over-polices and abuses minorities and the disadvantaged. Pushing for greater transparency and humane police practices hasn't worked here, either (which is also predictable from public choice theory).

    4. Don't you think you are jumping the gun a little on transparency? We have barely even gotten started.

      I agree with you about objecting because enforcing the policies requires miliitarization,. I'm just less persuaded that it requires discrimination and that there is not a darn thing we can do about it. Do you think we have made few substantial gains in the equitable treatment of minorities in these areas during our lifetimes?

      I think your position here is an awkward one, because, as I indicated in my first comment, the vast majority of statutes I'm happy we have (against rape, murder, theft, drunk driving, etc.) are not enforced in an even handed way. I'm not listening to anyone who wants these laws repealed for that reason, and I doubt you are either. I suspect we think that the good achieved by having laws like this justifies the predictable ways they will be abused, but if that is so we need to be open to the possibilty that gun control would be the same. Which makes the argument rest entirely on the empirical case for their effectiveness.

    5. But I tried to respond to that part of your first comment: the problems are much, much worse where it's difficult to identify a particular victim or a complainant, like laws that criminalize selling and possessing things. It's near impossible to enforce these sorts of laws in a non-invasive way. (For example, NYC's stop-and-frisk was explicitly justified as a gun control measure -- do you think there would have be as much political support for stop-and-frisk if the vast majority of people stopped were white?) Having the laws at all depends on them being enforced unfairly. That's how rape, murder, theft, etc. are different. Racial disparities in convictions for the latter sort of offenses are smaller and I suspect would be even smaller if you screened out offenses in those categories related to the drug war.

    6. Yes, that's an interesting point. I forgot about it. I am still just not sure how much significance to attach to it.

      First, there are many crimes in which it's difficult to identify a particular victim, such as DUI or even driving without a license that we are not interested in repealing on your argument. We can also just widen the canvas and ask whether you think you are making an argument that is strong enough to warrant repealing gun legislation that has long been in effect in other liberal democracies.

      Second, even if racial disparities in convictions are lower for things like rape, murder and theft, it does not follow that the resulting harm is lower since the severity of the respective punishments may differ. And I'm not sure what you say about convictions holds true of sentencing where there appears to be substantial disparities across the board.

      At the end of the day, though, I agree completely that when considering new legislation, the question whether it is likely to be another vehicle for harrassing disempowered minorities should loom large. It would just be a shame if we dismissed such legislation on that basis alone, especially if, as is quite possible, minorities would also stand to benefit most from it when it works.