Bryan Caplan  

Crusade, Denial, or Concerned Tolerance?: The Case of MSM and HIV

PRINT
George Mason University's Econ... The Curious Absence of Sociall...
Briggs and Tabarrok provide strong evidence that gun ownership increases suicide risk.  The response to their research provides strong evidence that readers swiftly twist research to suit their prejudices.  A slight caricature of the two standard reactions:

Reaction #1: Crusade.  Guns cause suicides, suicides are bad, so guns must be banned.  Anyone who disagrees is anti-science.

Reaction #2: Denial. Guns mustn't be banned, so either guns don't cause suicides, or suicides aren't bad.

The wise reaction, though, is what I call Concerned Tolerance.

Reaction #3: Concerned Tolerance.  Guns cause suicides, and suicides are bad.  But despite this risk, guns also have major upsides.  Self-defense aside, shooting and hunting are two popular American hobbies.  It's far from clear that the suicide risk outweighs these benefits.  But perhaps gun-owners underestimate this risk.  So let's politely publicize the Briggs-Tabarrok findings, then let individuals make their own choice.  In many cases, informed individuals will take extra precautions (e.g. locking up their guns) rather than actual quitting their risky behavior.

Concerned Tolerance may seem like mere libertarian dogma, but the truth is that most non-libertarians rely on Concerned Tolerance most of the time - even for choices that are vastly riskier than gun ownership. 

Like what?  Consider male-on-male sex.  It clearly raises your probability of dying from HIV.  By how much?  This impressive study by Susan Cochran and Vickie Mays (American Journal of Public Health, 2011) estimates the probability that men who have sex with men (MSM) will die of HIV-related causes.  Since it is based on past deaths, it is quite possible that the risk has changed.  But it's the best piece I could find on the topic, and you've got to start somewhere. 

The sample:
We used data from a retrospective cohort of 5574 men aged 17 to 59 years, first interviewed in the National Health and Nutrition Examination Survey III (NHANES III; 1988-1994) and then followed for mortality status up to 18 years later. We classified men into 3 groups: those reporting (1) any same-sex sexual partners (men who have sex with men [MSM]; n = 85), (2) only female sexual partners (n = 5292), and (3) no sexual partners (n = 197). Groups were then compared for all-cause mortality, HIV-related mortality, suicide-related mortality, and non-HIV-related mortality.
The risk estimate:
Compared with heterosexual men, MSM evidenced greater all-cause mortality. Approximately 13% of MSM died from HIV-related causes compared with 0.1% of men reporting only female partners. However, mortality risk from non-HIV-related causes, including suicide, was not elevated among MSM.
Of course, some MSM would have died of other causes during the sample period even if they never contracted HIV.  But very few.  If you assume that HIV-positive men would have died of other causes at the normal rate (6.6%), this still means that being an MSM raises your probability of HIV-related death by 12 percentage-points.

Converting that figure into an annual estimate (12 extra percentage points over 18 years) implies a .63 percentage-point increase in the annual risk of death by HIV.  That is an staggering risk.  By way of comparison, Briggs and Tabarrok found that People Living in Households with Guns (PLHG) had .01 percentage-point increase in their annual risk of death by suicide.  Thus, the risk of dying of HIV because you're an MSM is about 60 times greater than the risk of dying of suicide because you're a PLHG.  If another research design found that Cochran-Mays overestimated by a factor of ten, the Cochran-Mays effect would still be over six times as large as the Briggs-Tabarrok effect.

How should we react to Cochran-Mays' risk estimate?  As usual, we've got three basic options: Crusade, Denial, or Concerned Tolerance.

Reaction #1: Crusade.  Male-on-male sex is very dangerous, so let's pass and enforce draconian laws against it.  Anyone who disagrees is anti-science.

Reaction #2: Denial.  Male-on-male sex isn't dangerous.  Cochran-Mays must be incompetent and/or liars.  Or maybe male-on-male sex is dangerous, but people wouldn't engage in it unless the benefits outweighed the costs, so there's no need to study or publicize the risk.

Reaction #3: Concerned Tolerance.  Male-on-male sex is very dangerous, and people could easily underestimate its risks.  But despite this risk, male-on-male sex has major upsides.  Some guys really love it; some would feel spiritually incomplete if they quit.  So despite the massive risk of being an MSM, it's far from clear that this risk outweighs these benefits.  But perhaps MSM underestimate the riskiness of their behavior.  So let's politely publicize the Cochran-Mays findings, then let individuals make their own choice.  In many cases, informed individuals will take extra precautions (e.g. having fewer sexual partners) rather than actual quitting their risky behavior.

As far as I can tell, the vast majority of modern Americans take a Concerned Tolerance approach to MSM.  If they're right to do so here, why not across the board?

P.S. Please don't protest that "Sexual orientation is not a choice."  I'm talking about sexual behavior, not orientation.


Comments and Sharing





COMMENTS (25 to date)
Bryan H. writes:

Americans take the "ban it all" approach to things that are much safer than gun ownership. Ask the average American if he thinks anabolic steroids should legalized for possession. In fact I think the "incarcerate everyone who does it" approach is common across many areas.

In the end its not about what is statistically safer or not. Its about morality--whether a particular substance is believed to be good or evil. If a substance that statistically much safer than alcohol is believed to be evil, then voters will make it illegal. If "guns are bad" is the moral script in someone's mind, statistics do not matter.

All humans are moral beings, very few of them look at statistics on every issue.

[Nick changed to Bryan H. with commenter's permission, for clarity. --Econlib Ed.]

Brian writes:

Bryan,

You don't have to convince me about the Concerned Tolerance route--I'm already on board--but the other Bryan is correct that most people won't be swayed by statistics. Think about air travel or nuclear power.

On the other hand, I'm surprised by the stats you quote from the study. It looks like only 1.5% (85/5574) of men report engaging in ANY MSM. How can that be when the typical estimate of gays in the population is more like 3.5%? On top of that, a full 3.5% report NO sexual partners at all in the 17-year period? How can the unsexed outnumber the homosexed by more than 2 to 1?

Someone from the other side writes:

Disclaimer: Not engaging in MSM. Not a biologist or epidemiologist, either.

Since it is based on past deaths, it is quite possible that the risk has changed.
It would have changed dramatically: 1a) Mortality due to HIV has declined a lot [in developed countries] - modern medicine in fact is so good at controlling it that viral load can go below detectability. It is closer by now to a chronic than a fatal disease in developed countries (it apparently still lowers life expectancy but then so does diabetes). 1b) Some time ago a couple of European regulators were even publicly considering declaring sex with properly treated, low viral-load HIV positive partners as practically safer sex

Interestingly (but only tangentially related) reports are that gay porn (not interested enough to go do first hand research) has gone almost completely safer sex for quite a while now while straight porn decidedly has not. This may be related to the fact that transmission of HIV is far more likely with certain sex acts than with others.

Eric Hanneken writes:

Guns cause suicides, and suicides are bad.

Always? For everyone? While it's true that many suicides are impulsive, I'm not inclined to trust a blanket evaluation from an armchair over a person's self assessment.
Noah Yetter writes:

What if we believed "suicide isn't bad" before the Briggs-Tabarrok finding?

Suicide is a fundamental human right.

drycreekboy writes:

@Eric Hanneken

The meta issue isn't so much that guns cause suicides, but that they make a successful attempt more likely than a lot of other methods. You could drink a gallon of anti-freeze and kill yourself as reliably dead as with buckshot to the temple, but your gag-reflex gets in the way.

To the extent that many (by no means all) suicide attempts result from transient rather than chronic mental conditions then Briggs-Tabarrok make a good case we should look at whether density of firearm ownership means significantly more such people end up dead. According to the study, substitution effects don't make up the difference when there are fewer guns around, domestically anyway.

I hasten to add I'm a firm believer in gun rights, and one study does not by itself prove the connection, or its strength -- a qualifier I imagine Briggs and Tabbarok would share. A lot of questions need to be asked about that study, and I think the suggestion of reducing firearms ownership by ten percent at the end of the study is a pipe-dream in a country of ~300M private firearms. Just the amount of time involved in achieving such a goal might bring into play other effects bringing the suicide rate up or down. The number of lives in question could likely just as reliably be saved by raising the legal driving age to 18, or lowering the interstate speed limit back down to 55 -- good luck with either of those, and they'd be easier to implement. Nevertheless, what Briggs and Tabarrok are bringing to the table is qualitatively different from Brady/VPC/Bloomberg agitprop, and it deserves thoughtful consideration.

Eric Scheie writes:

In order to evaluate the risk properly, I think "men who have sex with men" should be broken down into safe, safer, unsafe, and maybe even very unsafe categories. It is important to remember that there are a number of sex practices that cannot transmit HIV. Which means that many "men who have sex with men" face zero to very low probability of ever dying from the disease.

Matt writes:

@Noah

I think you are badly mistaken, because I think you are trying to take arguments for assisted suicide and interject them into the gun control/suicide debate.

First, most people who try to commit suicide and fail usually don't try to do it again. This probably suggests that, yes, suicide is a bad thing for them, personally.

Second, if you are capable of killing yourself, whether it is a fundamental right or not is essentially irrelevant as there is no way to enforce a law against it. If you are permanently decided on killing yourself, you will get it done.

The fact is that most people that try to commit suicide are suffering from some mental disease/disorder that is often correctable. Suicide truly is "a permanent solution to a temporary problem."

DavoM4 writes:

"Guns cause suicides, and suicides are bad."

And ice-cream-laden spoons cause obesity, and obesity is bad. Gimme a break.

Suicides are caused by mental health problems. The only thing guns contribute to suicide deaths is that they make suicide attempts by mentally unstable persons more effective.

Eric Hanneken writes:

@drycreekboy

Actually, it was the second clause of Bryan's sentence I was contesting ("suicides are bad"), not the first ("Guns cause suicides"). I have no opinion on the quality of the Briggs & Tabarrok paper, although I think Bryan's three word summary implies that people who commit suicide have no agency, which is false.

Finch writes:

> I think Bryan's three word summary implies that
> people who commit suicide have no agency, which is
> false.

My reading on this suggests that suicide deaths are mostly a lot like drunk-driving deaths. There's some fault, but it's mostly an accident. A teenager gets broken-up with and really drunk and makes a terrible decision. A schizophrenia sufferer goes off meds or stays undiagnosed. That sort of thing. I don't think noble philosophers or euthanasia-candidates constitute a large fraction of suicides. It would make sense to offer reliable, painless methods to the latter categories and help to the former categories.

Jeff writes:

Most Americans, I would guess, view these two issues as fundamentally different. Gun ownership and use (whether for hunting, target shooting, or collecting) is a mere hobby, easily replaced with some safer, more wholesome activity like, say, shooting virtual people in video games. Sexual activity, on the other hand, is well-recognized as somewhat dangerous whatever your orientation, but it's also a fundamental animal behavior, like breathing or sleeping; thus it gets put in a different mental category. People are inclined to be more tolerant about it because it's something they do themselves (have sex...just not with same sex partners), whereas gun ownership to a lot of people seems dangerous, irresponsible, and, in many circles, unhip. Thus they either don't think about the benefits at all, or assign a very low value to it, even while recognizing others might disagree.

Eric Hanneken writes:

Finch,

Noble philosophers are indeed scarce, and yet billions of others manage to make decisions that have large effects on their lives: what career to pursue, whom to marry, etc. Remarkably, most of these people muddle through, although occasionally the results are disastrous. In spite of this imperfect record, I doubt Bryan or others on this thread would prefer their own judgement over a stranger's on the question of what that stranger ought to major in. The right answer depends on too many things that only he knows, including his own values and the "particular circumstances of time and place," to use Hayek's phrase.

But when it comes to a stranger's choice to commit suicide, apparently some of us know best, and the correct decision is almost always "Don't."

Don't get me wrong. If I were walking over a high bridge and saw someone about to jump, I would try to stop him, at least with words and maybe using force. Similarly, if I encountered someone on his way to throw away his high-paying career so he could be a drummer in a band, or someone who was about to leave his wife and marry an eighteen-year-old he met yesterday, I would try to stop him, at least with words (probably not using force). Those decisions have severe downsides, and shouldn't be made on impulse.

But do you know what I would do next? I would listen to the guy before I judged. Maybe that career was making him miserable. Maybe his wife was cheating on him. And maybe the would-be jumper's problems weren't so temporary after all. I would be willing to consider the possibility that the stranger was acting rationally. In one case at least, Bryan would not.

Motoko writes:

"I think "men who have sex with men" should be broken down into safe, safer, unsafe, and maybe even very unsafe categories. It is important to remember that there are a number of sex practices that cannot transmit HIV."

Well, this is certainly possible, but I seriously doubt the gun control crowd would allow us to break down categories of gun ownership into "safe, safer, and unsafe". Even if you try to make a category for someone who self-describes as locking their gun up away from children, the liberals will just argue that if they fail to do this just 5% of the time, it is unsafe. Similarly with MSM who self describe as having safe sex, one "abnormal" weekend and their risk factor is blown out of the water.

It kind of highlights the absurdity of all these regulations they want to put on guns - like require that they have biometric safeties and stuff - it won't end the complaints, because even if a method is 95% effective, the media will show us only the 5%. And no one was ever doing a cost-benefit analysis in the first place...

Finch writes:

Eric, I'm not entirely sure I understand your point.

If I could limit my own ability to make rash, irreversible decisions while intoxicated I would. Not that I'm ever really intoxicated, but I don't know what the future holds and I don't want to do something stupid on a really bad day. If I knew I would develop a mental illness, I would similarly, today, welcome constraints on my future actions.

I don't want to deny people choices, I just think most suicides don't really seem like choices, they seem like momentary lapses. I mostly agree with Bryan that this research does not make much of a case for gun control, though it could make a case for changes in personal behavior like storing your guns outside your home, or perhaps getting rid of your guns if you have been diagnosed with certain mental illnesses or have trouble with alcoholism or addiction.

By all means, offer medical euthanasia with a waiting period and counseling. Well thought-out desire to not be alive anymore isn't really a problem, and it doesn't seem to be a significant source of suicides.

Stud writes:

This makes similar points and touches on how to approach policy in an actually complicated world.

Noah Yetter writes:

@Matt

I draw no distinction between assisted and ordinary suicide, and I don't see how any of your objections are relevant.

Each of us has self-ownership, and to own a thing includes possessing the right to destroy it.

Suicide opponents want one thing only: to prevent other people, by force, from exercising control over their own lives. This I abhor above all else. It is irrelevant if their actions are motivated by genuine regard for the well-being of others or not. To forcibly prevent suicide is to make a man into a slave.

Brian writes:

"Each of us has self-ownership, and to own a thing includes possessing the right to destroy it."

Noah,

Ownership is not a good analogy for our own self-determination. As you say, we own "things," but people cannot be reduced to mere things, nor can we, even in principle, "use" ourselves.

You say "to own a thing includes possessing the right to destroy it." I don't think the law would agree. Especially when it comes to animals, which we CAN own, the law does not recognize a general right to destroy them. Ownership is really a protection against others doing something with the oblect, and not a right for us to do something with it.

You also say "Suicide opponents want one thing only: to prevent other people, by force, from exercising control over their own lives." And you know this HOW? Perhaps you should try to avoid slandering a large majority of the human race as being closet fascists.

The reality is that the result of successful suicide is irrevocable. Given that, any prudent person would discourage such an action and perhaps even employ modest coercion to prevent it until the reasons are better understood.

Finch writes:

"Suicide opponents want one thing only: to prevent other people, by force, from exercising control over their own lives."

Egad man, this is overreach.

Personally, I'd like it if we could distinguish the 14 year-old who turns out to be less popular than she imagined, the 16 year-old who finds himself gay, and the 19 year-old who really should have gone to a less selective college, from the guy with terminal cancer and the woman with the beginnings of Alzheimer's. We can reasonably protect the former category while making things easier for the latter. And unless you have some great ideas, I don't know of a fascist way to protect the kids with transient problems.

drycreekboy writes:

@Motoko

"Well, this is certainly possible, but I seriously doubt the gun control crowd would allow us to break down categories of gun ownership into "safe, safer, and unsafe". Even if you try to make a category for someone who self-describes as locking their gun up away from children, the liberals will just argue that if they fail to do this just 5% of the time, it is unsafe. Similarly with MSM who self describe as having safe sex, one "abnormal" weekend and their risk factor is blown out of the water."

I think this is spot-on, and dovetails well with what you see in the comments section in the MR postings on the study. A lot of people on the Left seem to believe no one can be trusted with a firearm unless they have the sanction of a uniform.

MingoV writes:

Gun ownership has not been proved to increase suicide risk. It is just as likely that suicidal thoughts increase the likelihood of gun ownership. A person who blew his brains out used a violent method to end his life, and that method indicates that he wasn't equivocating. A person who swallowed a bunch of pills may be equivocating (someone may find and save her), and the method isn't dramatic.

Mark V Anderson writes:

Funny how many of the comments have segued into a discussion of whether we should accept suicide.

I have to agree the most with Eric's last post. Yes, it is a good thing to try to discourage folks from taking rash actions, but in the end, it is the individual's decision. Several posters seem to imply that suicide is somehow different from other actions, that in this case we shouldn't allow individuals to make this decision. I disagree.

And I don't think it is that easy to kill yourself, at least not without enormous pain. I have never tried to kill myself, or considered it seriously, but I can imagine circumstances where I might well do it. If such circumstances do arise, I very much want to have a relatively simple, painless, and certain way to die, such as shooting myself in the head. I would dread being too feeble to take much action, and not even able to kill myself in my misery.

ThomasH writes:

I'd say the wise reaction is to recognize that guns in some circumstances promote/permit suicides and in other circumstances self defense. From that point one could ask if there are any changes that might be made in the way guns are currently regulated that would reduce suicides with little reduction of self defense. This however would be anathema to politicians who use allegiance to the NRA as a signal to a segment of the electorate.

It is nice to see a socially leftist blogger point out the difference between action and preference.

Jacob A. Geller writes:

I agree with the point of the post, but my goodness the MSM example was a poor choice. (See the 2nd & 3rd comments in this thread)

Why not just choose alcohol consumption? Almost everyone thinks prohibition is (/was) a bad idea. We're all proponents of Concerned Tolerance of alcohol consumption, at least above a certain age and outside the realm of operating heavy machinery.

Or sugar. Pick sugar. We're almost all proponents of Concerned Tolerance of sugar consumption.

...the problem, of course, is that almost any human activity you choose that exhibits some significant personal or social cost, is going to come along with some amount of support for some amount of regulation, large or small. We don't want 5-year-olds to drink; we don't want adults to drive drunk; but we *do* want alcohol to be legal for most people and under most circumstances; etc. etc.

What are we going to do about the fact that guns seem to "cause" some suicides (and murders, for that matter), on the margin? (Those of you who missed the key words "some" and "on the margin" in the previous sentence, please re-read it.)

Calling for Concerned Tolerance is an improvement over Crusade and Denial, but there's a great big grey area, and most people are in that grey area: Concerned Tolerance is a great starting point, but what should be the *limits* of Concerned Tolerance? Does having a child, for example, have any bearing on whether you should be allowed to own a gun, or on what circumstances you should be legally allowed to own a gun, or under what conditions you may keep a gun you own in the home, etc.? How about the presence of a mentally deranged individual living in your household (think Adam Lanza...)? Etc. etc. You can come up with many questions like this in the gray area between Concern/Denial and Concerned Tolerance.

My point is you've made a correct but unfortunately easy point. The devil is in the details of what the limits of Concerned Tolerance should look like. And when I look at the national conversation over gun control, that is where I see the conversation is mostly at: Most people are in the center, i.e. they want some limited but reasonable amount/type of regulation (even many, or even most, NRA members feel this way). You're really only addressing extreme fringes on the left and right, who want total gun bans and total freedom to bear every type of arm, respectively. But these are minorities.

And btw, even if most people were to *immediately* jump to Caricature #1 and Caricature #2, upon further reflection they abandon both Caricatures and settle rather quickly into Concerned Tolerance.

Comments for this entry have been closed
Return to top