How Economists Figure Sex

From a Wash Times review of economist (and occasional Reason contributor) Steven Landsburg's new book, More Sex is Safer Sex:

The "More Sex" thesis: If prudes occasionally slept with strangers, it would slow the spread of STDs.

Here's how it works. One such prude walks into a bar, and he's uninfected. If he takes home an uninfected woman, great -- he distracted her from a potential disease carrier. If he gets herpes, that's also great, because he's sexually conservative and won't pass the infection along very often. Better him than someone with less self control.

Either way, society benefits when the chaste open up slightly.

The reviewer concludes, the "book reads fast and provides plenty of intellectual red meat." Whole thing here.

Comments

Lupito41 | April 24, 2007, 10:32am | #

Seriously? Society benefits more than, say, if the loose were to tighten up? I know he doesn't spell out the dichotomy, but the implication seems clear.

robc | April 24, 2007, 10:32am | #

Im going to ask the counter question: If "prudes" can increase their partners up to 2.25 per year and decrease the spread of disease, how much less spread of disease would there be if the "less prude" decrease their # of partners to 2.25 per year?

Hmmm...as an aside, assuming a start at age 16 (age of consent in my state) @ 2.25 partners per year, almost 22 years for me = 49.5 partners. Even realizing that I need to count partners who cross year boundaries multiple times, I guess I must fall into the uber-prude category.

robc | April 24, 2007, 10:33am | #

Lupito41,

nice simulpost.

Lupito41 | April 24, 2007, 10:35am | #

robc,

I guess us sexually underfed have itchy trigger fingers.

Cab | April 24, 2007, 10:35am | #

robc - you owe it to me to go out and have sex with strangers more often.

Warren | April 24, 2007, 10:37am | #

While I've no doubt that more sex leads to a healthier society. I'm not quite buying the whole 'combating the spread of STDs by giving them to the "right" people'.

My theory is that the occasional collapse of prude willpower contributes to spreading STDs. My thinking is, a prude giving in to passion isn't practicing safe sex. The same unwillingness to deal with reality also keeps them from getting timely treatment. Just a theory.

But anyway, More Sex! whoo hoo!

robc | April 24, 2007, 10:37am | #

Cab,

Im going to stick with self interest on this on. But then again, Im a greedy libertarian.

Ken Shultz | April 24, 2007, 10:38am | #

"If he gets herpes, that's also great, because he's sexually conservative and won't pass the infection along very often. Better him than someone with less self control."

MikeT | April 24, 2007, 10:39am | #

Replace herpes with AIDS and see how it sounds...

What you are encouraging people to do is engage in risky behavior by hooking up with strangers. Society benefits when people at worst hook up with people they know and trust, who are probably likely to be open with them. Society benefits even more when people don't engage in behavior that causes unwanted pregnancies and risks disease in the first place.

If you want to hook up with someone, just do it. Only losers need socio-economic-political justifications for behaving that way.

mllh | April 24, 2007, 10:39am | #

robc

what is your state again? 16? hmmmmm

Ken Shultz | April 24, 2007, 10:39am | #

I think you have to be something of an intellectual to fall for an argument like that.

robc | April 24, 2007, 10:41am | #

mllh,

KY, but most US states are 16. Check out ageofconsent.com, I cant believe I knew the url for that. What kind of prude am I?

NaG | April 24, 2007, 10:41am | #

Uh, if "prudes" started sleeping around more, then they wouldn't be "prudes" anymore, now would they?

I sure hope Landsburg's thesis is better than reviewer's take, because it is self-evident that you can't take advantage of "sexually conservative" behavior when you're insisting that people be less sexually conservative. In fact, logically, Landsburg's thesis can only result in the exact opposite conclusion, that prudes acting less prudish will only increase the spread of STDs since a formerly sexually-conservative group that has a low incidence of STDs will be greatly increasing their risk.

ktc2 | April 24, 2007, 10:42am | #

And again, the solution is MORE SEX!

We need to all start behaving like bonobos with broadband.

John | April 24, 2007, 10:46am | #

"Suppose you walk into a bar and find four potential sex partners. Two are highly promiscuous; the others venture out only once a year. The promiscuous ones are, of course, more likely to be HIV-positive. That gives you a 50-50 chance of finding a relatively safe match.

But suppose all once-a-year revelers could be transformed into twice-a-year revelers. Then, on any given night, you'd run into twice as many of them. Those two promiscuous bar patrons would be outnumbered by four of their more cautious rivals. Your odds of a relatively safe match just went up from 50-50 to four out of six."

What he says makes perfect sense. It is a good example of how economic analysis can lead to counter intiutive truths. Landsburg is a really brilliant guy and his book is great. Much better than the more hyped Freakonimics. Landsburg is what the guy who wrote Freakonomics thinks he is.

joe | April 24, 2007, 10:48am | #

"One such prude walks into a bar..."

What is this, a joke?

thoreau | April 24, 2007, 10:48am | #

I think you have to be something of an intellectual to fall for an argument like that.

Ken-

That's kind of insulting.

As far as the point being made, I suspect that it's mathematically sound, but there's more to life than math. I know that a lot of work has been done on modeling the rate at which diseases spread through populations, and it wouldn't surprise me if in certain circumstances a disease spreads more slowly when the carriers give it to people who are less likely to spread it.

But as soon as you go from the population view to the individual view, it's advantageous for the individual to not get the disease.

To go back to the mathematical side of this, I don't see how "prudes" catching a disease that they won't spread is really a Nash equilibrium.

The Unknown Comic | April 24, 2007, 10:48am | #

I can see the public service ads now:

"Start to spread to stop the spread."

John | April 24, 2007, 10:50am | #

Thoreau they have done that research and it indicates that AIDS would spread more slowly if a larger number of people would take multiple partners.

That's why increased activity by sexual conservatives can slow down the rate of infection and reduce the prevalence of AIDS. In fact, according to Professor Michael Kremer of MIT's economics department, the spread of AIDS in England could plausibly be retarded if everyone with fewer than about 2.25 partners per year were to take additional partners more frequently. That covers three-quarters of British heterosexuals between the ages of 18 and 45. (Much of this column is inspired by Professor Kremer's research."

http://www.slate.com/id/2033/

mllh | April 24, 2007, 10:51am | #

Kentucky. I've never really thought much of Kentucky, until now. Horses, bluegrass, coal, smoking, whiskey, teenagers who can legally have sex: now that I think about it, LONG LIVE KY!

ooooh, even the state abbreviation is great!

robc | April 24, 2007, 10:53am | #

John,

The problem with that math is, as a "prude", my once a year night had a 1/2 chance of hooking up with an unsafe match. Now, with the adjustment, only 1/3 are unsafe. But, with 2 outings a year, my unsafe sex chance is now 5/9.

Lets assume the promiscuous ones are picking up partners in the bar once a month (vs once a year). By cutting down to 6 times a year, they have changed the results in the exact same way. There will now be twice as many "clean" as "unclean" partner options. This helps both the promiscuous and the prudes. The prudes get reduced to 1/3 chance from 1/2 for their once yearly excursion and the clean rate is better for the promiscuous too.

The first way may help OVERALL at the expense of the prudes, but the 2nd helps both groups.

thoreau | April 24, 2007, 10:55am | #

Oh, I'm sure that from a population standpoint this model is correct under certain conditions. I just don't see how it's a Nash equilibrium. Partnering with a carrier might stop him from spreading it to multiple people, so the total infection rate for society will slow. But I don't see how partnering with a carrier is a Nash equilibrium, i.e. it's always better for the individual to not partner with a carrier.

Maybe it becomes a Nash equilibrium under conditions of imperfect information, and where you include pleasure as part of the payoff calculation.

Disclaimer: I have no formal training in game theory, just tidbits that I've picked up here and there.

robc | April 24, 2007, 10:57am | #

mllh,

The smoking thing in KY isnt as strong as it use to be. Both Louisville and Lexington have smoking bans. As of July, the Louisville one expands to bars (previously exempted). The only business not covered will now be ... wait for it ... Churchill Downs.

Horseracing still has a lot of pull.

John | April 24, 2007, 10:57am | #

Robc,

That may mean that there is an "ideal" amount of sex that most restricts the spread of disease. That is an interesting point. Whatever the curve is though, its ideal is not "0", so I think Lansburg's indictment of abstenence is still valid.

mllh | April 24, 2007, 10:58am | #

"But as soon as you go from the population view to the individual view, it's advantageous for the individual to not get the disease."

That dichotomy between individual and population is the problem with politics/economics/lib philosophy/what-have-you, I think. It is best for the individual to avoid multiple sex partners, while it is simultaneously best for the whole country to encourage sleeping around. Abstinence is best for you; sluttiness is best for all of us.

Can economists/philosophers ever reconcile the two?

But, anyway, speaking of the tremendously glorious and sublime state of Kentucky...

robc | April 24, 2007, 11:00am | #

John,

If all we cared about was disease, the ideal point would be 0.

John | April 24, 2007, 11:00am | #

Of course abstinance is only best if you can really follow it. If you break down once in a while you could be in worse shape than if you were a slut.

John | April 24, 2007, 11:02am | #

"If all we cared about was disease, the ideal point would be 0."

But you have to work under the assumption that people are going to have sex. Further, the ideal point can't be zero because we would all die off if there was no sex. Clearly, people are going to be having sex. Given that fact and the fact that there are STDs, what is better from the standpoint of STD control, monogomy or multiple partners?

NaG | April 24, 2007, 11:02am | #

John: The flaw in the logic, of course, is that it doesn't take into account the fact that if prudes are going out more, they'll be taking more chances, and thus increasing their chances at being infected. In other words, yes, increasing the number of prudes in the bars will reduce your one-time shot of getting an STD. However, this is not a one-time analysis. If I am a prude, going out once a year, and face a 50-50 shot at infection, then I might think I'm getting a deal when us prudes now go out twice a year and my chance of catching an STD on a given night is now only 1 in 3. However, I'm forgetting that in order for this to be true, I now have to be in the bars twice a year instead of only once, which means I am now taking a 1:3 chance TWICE a year instead of a 1:2 chance once a year. Do the math and tell me which is better.

robc | April 24, 2007, 11:03am | #

John,

sure, but if everyone was a prude with the occassionally lapse, it would be better than a mix of prudes w/TOL and sluts.

Queen's Gambit Declined | April 24, 2007, 11:03am | #

I think the element missing here is that there are plenty of people who would like to have more partners but are unsuccessful at finding them.

John | April 24, 2007, 11:03am | #

The problem with democracy is not that politicians kowtow to financiers and lobbyists; it's that politicians kowtow to their own consituents, spending other people's money along the way. In other words, their incentives are all wrong. Effective reform should supply better incentives.

So if I could make just one change in the American political system, it would be to give each voter two votes in every congressional election. You'd get one vote to cast in your own district and another to cast in the district of your choice. When a congressman from West Virginia funnels taxpayers' money from fifty states to his home district, I want him to face the prospect that taxpayers from fifty states will share their feelings with him on election day.

I'd also redraw the boundaries of Congressional districts according to the alphabet instead of geography. Instead of congressmen from central Delaware and northern Colorado, we'd have a congressman for everyone whose name begins with AA through AE, another for everyone whose name begins with AF through AL, and so on. The point being that it's easy to devise a pork barrel project that benefits everyone in northern Colorado, but a lot harder to devise a pork barrel project that benefits everyone whose name happens to begin with Q.

Finally, I want federal income tax rates determined separately in each congressional district, as a function of how much spending your congressman has voted for. The more he votes to spend, the more you pay in taxes. That should solve the problem of voters who pay little attention to what their representatives are up to.

If you're worried about this deterring congressmen from voting for bills that are truly in the national interest, I'm willing to make an exemption for any spending bill that passes by a supermajority of, say, 70%.

Am I serious? Of course I'm serious. Of course I'm also aware that our legal system would probably render any of these reforms quite impossible, and that this of all blogs is the one where readers will jump in to tell me why. But the disconnect between congressional incentives and the welfare of the general public is real, and needs highlighting. So when I say "Let's redraw the congressional districts according to the alphabet", what I'm really saying is "Let's think hard---and creatively---about ways to sever the link between parochial interests and congressional incentives." That's an entirely serious point.


http://volokh.com/archives/archive_2007_04_22-2007_04_28.shtml#1177387121

I kind of like this idea to.

Warren | April 24, 2007, 11:03am | #

A prude walks into a bar.
The bartender says "what'll ya have"
The prude says "I'd like a Virgin Bloody Mary".
Then the gal on the stool next to him pipes up "There was just the one time, I should start menstruating next week, and the name's Teresa. But I'm pleased to meet you. What do folks call you?" Then the prude says, get this, "They call me the Virgin Rob Roy"
Get it? The Virgin Rob Roy Har Dee Harr Harr

I've got a million of em

robc | April 24, 2007, 11:04am | #

NaG,

I did that math in the post at 10:53.

Ken Shultz | April 24, 2007, 11:04am | #

"That's kind of insulting."

It wasn't meant to be. ...but there are certain pitfalls that a prudish, bible thumping farm worker is unlikely to fall for.

Even we assume that society does get better when certain individuals contract STDs...

Why should he go out and get herpes for the betterment of society? What's in it for him? Why is unprotected sex with strangers in his best interest? Why would he, in his culture, in his community, want to do that?

Is the realization that society is better off now that he's infected supposed to be something he consoles his wife with after he's passed it on? Again, assuming his wife isn't an intellectual, I'm not sure she's going to appreciate the value of that pearl of wisdom.

John | April 24, 2007, 11:05am | #

"sure, but if everyone was a prude with the occassionally lapse, it would be better than a mix of prudes w/TOL and sluts."

But sluts happen. You can't eliminate all sluts anymore than you could eliminate all prudes. The issue is which side of the scale should we err towards.

NaG | April 24, 2007, 11:06am | #

Of course, robc already beat me to the punch. Good job.

The stunning thing about Landsburg's thesis is that it happens to be flat wrong on a very basic level: it forgets to apply the change in rules to everyone. Sorta reminds me of the basic premise of Hazlitt's "Economics In One Lesson." How could an economist like Landsburg make such an elementary mistake?

John | April 24, 2007, 11:06am | #

"Even we assume that society does get better when certain individuals contract STDs..."

You miss the point. The point is that fewer people would be infected. See the post above regard the spread of AIDS in England.

thoreau | April 24, 2007, 11:07am | #

Is the realization that society is better off now that he's infected supposed to be something he consoles his wife with after he's passed it on?

I think the point isn't that society's better that he's infected, but rather that society's better because he may have saved multiple others from being infected.

But you're right, how does that console him? Answer: It doesn't. Which is why I don't see how this is an equilibrium strategy.

Unless it's changed by imperfect information on disease carriers, and non-zero utility from sex.

robc | April 24, 2007, 11:08am | #

John,

Im going to fall on the side of letting the individual decide. I will stay in the prude camp. If that mean more sluts get diseases because Im not out there protecting them, then so be it.

BTW, this is kinda fun using prude and slut as technical economic terms.

Call Any Vegetable | April 24, 2007, 11:11am | #

looks like a commie style public health plan

one that's geared for a 5 decade run

none for me, thanks

if you lack the self control to keep from banging a (newly met) diseased person, and end up catching something, that's your problem, not mine

I practice real safe sex:

I don't fuck anything nasty

suggesting things would be better for all involved if only I were to imitate the promiscuous sounds like a call for cannon fodder

you know, in order to spare the "real" soldiers

LarryA | April 24, 2007, 11:11am | #

Unintended consequence department:

"One such prude walks into a bar..."

finds a partner, takes her home, they have sex.

The next day he looks in the mirror and says, "Gee whiz, that was okay. I think I'll go ahead and get my other sacrifice over with for the year.” That night he walks into a bar, picks up a different woman, takes her home, and has sex.

The next day he looks in the mirror and says, “By god, that was fun too. It’s really 2.25 partners per year, so I’ll go the extra mile and use up the fraction. Save some other guy from having to do it.” That night he walks into a bar, picks up a different woman, takes her home, and has sex.

The next day he looks in the mirror and says, “Hot damn, this sex stuff is alright. It’s only six months to the end of the year, so I’ll just work ahead a bit.” That night he walks into a bar, picks up a different woman, takes her home, and has sex.

The next day he looks in the mirror and says, “Screw the study, I’m tired of being a prude. Go for it.”

John | April 24, 2007, 11:12am | #

Robc,

It also means that your foreys into the slut camp are going to be risky. I guess the lesson is that if you are going to be a prude, really be one.

Lansburg makes the point that the problem is that people who are disease free are not properly rewarded because of inperfect information. If we knew who had diseases or was less likely to have them, those people would be more in demand for sex. Bascially, the disease free people would have a lot sex with each other instead of risking it with the diseased. He admits, that it would be hard to have such a system in reality, but for now, free condoms and an end to abstinence campaigns would help.

mllh | April 24, 2007, 11:16am | #

I'm way out of my intelligence comfort zone, but it occurs to me that evolution could also be at work here. Suppose it's best for our species as it currently stands for everyone to be a slut (as the thesis suggests). Yet there are among us prudes. Will STDs select prudes for survival, making future generations more and more prudish? In other words, is prudishness in the presence of widespread STDs an evolutionary-adaptive strategy?

robc | April 24, 2007, 11:18am | #

John,

Yep, best if prudes only have sex with other prudes. The incredibly prudish idea of not sleeping with strangers handles some of the imperfect information problem. With a proper vetting period, most (but not all) of the risk can be reduced.

NaG | April 24, 2007, 11:23am | #

Of course forays into the Slut camp will be risky. The problem is that instead of increasing information so that it is easier to find non-infected partners, Landsberg proposes more forays into danger. His proposed solution is exactly what would make the problem worse -- or at least the solution as currently described by the reviewer. Maybe he will have some nuances when he posts on it on Volokh that might help.

One thing that is curious to me is how pornstars, who are on the high-end of promiscuity, tend to be the most vigilant in protecting against STDs. Why? Because they know that they bear an increased risk by the nature of their job, so they take control of the situation by demanding information. If one pornstar winds up being diagnosed with AIDS, suddenly everyone who ever worked with that star is suddenly in question and a danger. So, knowing this, why don't non-professional members of the Slut camp do the same? If you know that your promiscuity increases your risk, you should be much more insistent on information to protect yourself. However, this does not tend to happen. Why? My guess is, no one wants to think of themselves as members of the Slut camp. And that may be the bigger problem.

carrick | April 24, 2007, 11:25am | #

The article at Slate that John reference is one of an ongoing series of economics and how it effects everyday life.

The common thread in the series is that a rational, statistical analysis of every day activities leads to dramatically different, and frequently counter-intuitive, results than might be expected by the average joe or jane on the street.

In this specific case, the population as a whole would benefit if the effects of promiscuous people in the population were diluted by more non-promiscuous people engaging in "loose sex". Note that the outcomes for specific individuals may be far from desireable, but the population as a whole benefits.

This topic should resonate with libertarians who normally take the position that it is the rights of individuals that takes precedence over the well-being of the population.

dhex | April 24, 2007, 11:27am | #

prudishness for the prudes.
sodom for the sodomites.

everyone's happy!

(if only this were true)

that said, the book seems kind of insane. or it assume certain patterns of sexual behavior are really, really predictable?

mllh | April 24, 2007, 11:32am | #

Can private enterprise get involved in this?

I have an idea for a "Slut Club". Members could pay a quarterly fee for certification. Certification would involve a regular physical exam and test for all common STDs. Those who pass the exam would receive a "license" which would certify them clean and available for sex. This license could be presented to all potential partners over drinks; the last convincing argument for why "you should come back to my place." Naturally, sluts with such certificates would prefer other sluts with similar certification.

If this enterprise were to succeed, then I could make millions when my purely voluntary certification became the gold standard for choosing a fellow slut for the night.

carrick | April 24, 2007, 11:34am | #

The other point of this article is that the "market" for "loose sex" is broken, because individuals do not have access to the necessary information to determine the "value" of any individual partner.

If I could advertise that I am known to be disease-free, I can command a better price for myself if I choose to put myself on the market.

If I could see the advertisements for available partners I could also evaluate what I am willing to spend to get a specific individual.

Since these advertisements don't exist, we could flood the market with disease-free individuals and reduce the likelyhood that any given transaction would have a negative outcome.

Borat | April 24, 2007, 11:34am | #

I like sex!

It's nice.

robc | April 24, 2007, 11:35am | #

mllh,

Go for it. It worked for Underwriters Labs (except for the making millions part, I dont know if they make a profit or not).

Im not great at marketing, but you might want to call it something other than "Slut Club".

robc | April 24, 2007, 11:36am | #

carrick,

The market isnt broken, its merely providing arbitrage opportunities.

The Wine Commonsewer | April 24, 2007, 11:36am | #

Landsberg reminds me of the guy locked up in a cell in the looney bin who was balancing a cashew on the end of his stiffie. When asked why he was locked up the patient replied.....

Isn't it obvious? I'm fucking nuts.

NaG | April 24, 2007, 11:40am | #

carrick: In this specific case, the population as a whole would benefit if the effects of promiscuous people in the population were diluted by more non-promiscuous people engaging in "loose sex".

No, that's exactly wrong, for the reasons stated above. The dilution would be temporary at best (depending on the ratio of prudes to sluts), and would inevitably lead to higher incidences of STD infection. The reason why prudes have low incidences of STDs is BECAUSE they are prudes. Landsburg cannot spread this benefit by subverting the very behavior that caused the benefit in the first place.

NaG | April 24, 2007, 11:41am | #

Also, carrick, people advertise being STD-free in personals ads all the time. However, personals ads are not beacons of truth. A third-party verification procedure, like what mllh suggests, is the best approach. That's what pornstars do.

mllh | April 24, 2007, 11:42am | #

Slut Club was MY idea! Mine! No one may copy it! I'm going to incorporate and patent and everything else right now!

(Suggestions for alternate names are welcome. Best suggestion gets to be my business partner.)

jtuf | April 24, 2007, 11:48am | #

If you don't assume random mating, the results are a bit differret. Say prudes only mate with other prudes. They benefit from a much lower disease rate. The non-prudes would have a higher infection rate, which could cause epidemics to burn out more quickly.

carrick | April 24, 2007, 11:49am | #

No, that's exactly wrong, for the reasons stated above. The dilution would be temporary at best (depending on the ratio of prudes to sluts), and would inevitably lead to higher incidences of STD infection.

The study in England comes to a different conclusion.

Unfortunately, I am not deeply trained in statistics or the transmission of communicable diseases. So I can't confirm or contradict the reported study. But the results of the study don't surprise me that much.

And NaG, your analysiss smacks to much of "moral outrage" to carry much weight in contradicting the reported results.

carrick | April 24, 2007, 12:01pm | #

mllh, try Sluts "r" Us

You need more than a certificate to be successful. Since it takes some period of time after infection for a test to show that infection, you will need to track every encounter so that 1) individual can see every transaction that occurred since the last test and 2) potentially infected patrons can be tracked down quickly if someone "breaks the rules" and has a transaction outside the system.

NaG | April 24, 2007, 12:06pm | #

Nope, carrick, no "moral outrage" here. Nothing I've said could be implied as such -- se my comments on pornstars, for example. Landsburg simply wants what he cannot have: prudes who don't act like prudes. The logical contradiction is quite simple.

The study in England does NOT come to a different conclusion. The best it can say is that there is a "plausible" benefit, which is far, far, FAR from a real measured benefit. Landsburg's poor logic has apparently suckered some people with its "plausibility," until the above contradiction is pointed out.

Me, I think everyone, prudes and sluts, should all demand rigorous information from every sex partner. Third-party verification sources should proliferate. After all, STDs don't care if you're a prude or a slut when infecting you, so be careful no matter what.

Mackie | April 24, 2007, 12:12pm | #

A prude that gets an STD the first time out will have his prudity reinforced, resulting in a bitterly regretful person who's not getting any. Ergo, a net deficit for society.

carrick | April 24, 2007, 12:23pm | #

This is from memory, so there may be some errors . . .

STDs are generally difficult to transmit, hence the requirement for intimate contact. The probability of infection is well below 50/50 for any given encounter. Infections are typically the result of repeated encounters with infected partners.

The nature of the encounter also influences the probability of transmission -- women are far more likely to get HIV from an infected man than the other way around.

So assuming a low percentage of the pouplation is infected and a low probablity of getting infected from a single encounter, then random sex in a large population flooded with disease-free partners could well reduce overall transmission rates.

That's not a proof, only a hypothesis.

Brandybuck | April 24, 2007, 1:31pm | #

The biological math doesnt't add up. All you've done is to infect 2.25% of prudes with STDs. And because you are encouraging them to have promiscuous sex, just at a lower rate, they will continue infect others. Societal STD levels would be lower if the monogamous stayed monogamous.

Human beings are not fruit flies. This isn't the same as releasing some sterilized male flies into the environment. Human beings are always "in heat" and can mate as often as they wish. Would 2.25% of those disatisfied with a prude's sexual peformance (they're not experienced, you know) go back to that bar to try for another round the same night? Sounds plausible.

NaG | April 24, 2007, 1:53pm | #

carrick: "So assuming a low percentage of the pouplation is infected and a low probablity of getting infected from a single encounter, then random sex in a large population flooded with disease-free partners could well reduce overall transmission rates."

How? Again, more exposures at lower percentages can easily result in higher chances of transmission, as demonstrated in the calculations offered earlier today. Doesn't matter whether the lower percentages are due to the ability to transmit infection or ability to find a clean partner -- the more often you take chances, the more likely you will get burned in the long run.

Let's take this a step further. Going with the 50-50 chance that Landsberg likes, let's say the typical bar today has 50 prudes and 50 sluts on a given night (not considering gender). You walk in, and you have a 50-50 chance of going home with an infected partner, as do all the other prudes. However, each slut has a 51-49 chance of going home with a clean partner (you count as one of the 50 prudes). Thus, while you have a 50-50 chance of not catching a disease, each slut has slightly better than 50-50 chance of spreading it.

Now, let's let all the prudes in. Now the given bar has 500 prudes and the same 50 sluts. You walk in. Now all the uninfected bargoers have 10:1 odds of going home with a clean partner, which is much lower than before. Great, huh? Except now every slut has 501:49 chance of getting a clean partner too, meaning now they have better than 10:1 odds of spreading their STD. Assuming that it's not a sure thing that every slut finds a partner every night, adding in all the prudes has greatly increased the chance of each slut hooking up for the night. So not only are more sluts hooking up, but they are far more likely to hook up with uninfected partners, too.

What we've done is lower any one prude's one-time chance at getting an STD, but greatly increased the chance that an STD carrier will spread the disease. That means we have spread out the risk but yet increased the overall risk. There is no way to look at this without concluding that overall transmission rates would go up as a result. It's a classic case of mistaking the forest for the trees.

stephen the goldberger | April 24, 2007, 2:09pm | #

i studied under landsberg at the university of rochester. His classes were exactly like his books. Fascinating and provactive arguments within the framework of microeconomic reasoning. I recomend them wholly.

| April 24, 2007, 2:49pm | #

What do prunes have to do with it?

a | April 24, 2007, 3:02pm | #

What we've done is lower any one prude's one-time chance at getting an STD, but greatly increased the chance that an STD carrier will spread the disease. That means we have spread out the risk but yet increased the overall risk. There is no way to look at this without concluding that overall transmission rates would go up as a result. It's a classic case of mistaking the forest for the trees.


You should really check out one of Michael Kremer's papers (alluded to above) on this topic, e.g. this one. This phenomenon is interesting precisely because it's so paradoxical.

Basically what happens is that under the right conditions, the average number of partners for an infected person goes down (as more prudes are infected), which in turn reduces the transmission rate despite the increasing number of sexual encounters. This is a dynamic problem, so you have to look at more than one time period, and the effect is not always observed, as it depends on the initial conditions and parameters of the model. But it is nevertheless true in some cases.

jadagul | April 24, 2007, 3:21pm | #

NaG: First of all, not all sluts are infected. That's the main flaw in your argument: the major benefit of increased sex by prudes is the very fact that it makes prudes more likely to be infected—when a disease carrier infects someone who has sex twice a year, much less damage happens than when he infects someone who has sex fifty times a year. The goal is to have a higher proportion of the infected be prudes and a smaller proportion be sluts, because infected sluts can do so much damage.

To respond to other objections: the idea isn't to turn prudes into sluts: if in the status quo, sluts have 50 partners a year and prudes 1, Landsburg wants to change the numbers to 50 and 2. The idea is that if x is the number of partners each prude has each year (holding the number of partners for each slut constant), infection rates are minimized for some value of x, and this value is non-zero. Obviously it's possible to overshoot the optimal value, and x=50 is probably worse than x=0. But x=2.25 is apparently even better (under plausible modeling conditions; I haven't read the paper myself, so I don't know what the conditions are. But they're probably no sillier than the assumptions of most powerful and useful microeconomic models).

Finally, no one is advocating forcing the prudes to have sex. The point is that we devote some resources to encouraging people not to have sex; the main impact of these campaigns is to deter prudes from having sex. But more sex from prudes actually benefits society overall, even if it is a net harm to the prudes. So we should stop trying to discourage prudes from having sex. In fact, his argument is precisely that it benefits society but hurts the prudes for the prudes to have more sex; this is why he wishes to compensate prudes for having sex to encourage them to have more of it. So, Thoreau, his point is that more prude sex isn't (necessarily) a Nash, but that we can institute side payments to make it both Nash and Pareto superior.

NaG | April 24, 2007, 3:31pm | #

a, the problem is that Kremer's paper is full of potentials and hopes but no real concrete results. It's all "preliminary calculations" and so forth. Here's the best face I can put on Kremer's thesis, going with the prude/slut terms we've been using so far: Prudes, by taking themselves out of the dating pool, make it more likely for sluts to have sex with infected partners. If sluts can have sex with prudes instead of fellow sluts, then they will be less likely to contract STDs and, by their promiscuity, then transmit that STD to others. Am I being fair, here?

The problem, yet again, is that this invariably increases the potential of each infected slut to transmit their STD to an uninfected partner. There's no other way to cut it that I can see.

Also, once you start hedging on "right conditions" when talking about the hormonal dating scene, I think the premise gets thrown out the window.

I agree with Kremer that information and condoms are better than abstinence programs. But I think his thesis is more provocative than illuminative.

NaG | April 24, 2007, 3:46pm | #

jadagul: I know that not every slut is infected. I'm just going with Landsburg's own 50-50 model.

First, the argument that prudes having more sex doesn't make them non-prudes is specious. It defeats the point of there being prudes in the first place. But anyway, it is correct to say that an infected slut who has, say, 15 partners in a year can do far more damage than an infected prude who has only one partner in a year. However, with the prudes in the dating pool, the average infected slut is likely to infect more people out of his 15 than before, when he was more likely to be having sex with fellow infected sluts. Last I checked, measuring transmission rates does not depend on how many sexual partners that victim will have. Each additional infected person counts equally. Thus, when prudes jump in the pool, infected sluts are far more likely to have sex with uninfected partners than before, and thus will spread the disease faster. That their victims might not spread the disease as fast themselves is of small consolation now that we realize that we've vastly increased the potential pool of victims from what it was before. The ceiling has been blown open. If you're lucky, you have fewer infected sluts to spread the disease, but they are far more successful at doing so. This is a gain?

jadagul | April 24, 2007, 4:43pm | #

NaG:

Well, first, it's also unlikely to be anywhere near a 50-50 split of infected/uninfected sluts. The socially optimal number of partners per prude is a parameter that depends on the other parameters you plug into the model; it's entirely possible that if we assume 50-50 the socially optimal outcome is for the prudes to abstain entirely until the sluts all kill themselves off (it's also possible those rates would make more prudes even more desirable—I haven't done the math, so I don't know the direction of the effect). Kremer's paper, from what I can tell, says basically that if you plug in the actual numbers from British AIDS infections, the ideal number is about two an a quarter, which is larger than the actual number. So under the conditions that actually occur, it would be socially desirable for prudes to have more sex, by a reasonably small amount.

And the argument that prudes having more sex doesn't make them non-prudes is specious only for unreasonably large effects. If we move from 50 partners a year and 1 partner a year to 50 and 3, there's still a pretty clear split.

NaG | April 24, 2007, 6:04pm | #

Except for the fact that one does not immediately go from "prude" to "slut." It's more that one goes from "prude" to "not-prude." Perhaps "average." The whole point of prudes is that Landsberg is seeking to harness their sexual conservatism as a mechanism for restraining the spread of STDs, but he is doing so by asking that they be less sexually conservative! He is inherently weakening the very behavior he seeks to gain from.

I don't know what the incidence of STDs are among sluts, or prudes for that matter. Landsberg used the 50-50 model to prove his point, so I figured it was fair game for me to use the same model to prove mine.

a | April 24, 2007, 8:14pm | #

Like I said above, this paradox is sensitive to the parameters of the model you choose and the initial conditions, but here's an example I concocted that works.

Suppose that there are 10000 prudes and 1000 sluts, and that no prudes are infected to start, while 10% of sluts are.

Next, suppose that sluts seek a mate at every time period, while prudes seek a mate with probability p. Furthermore, assume mating is random among those seeking mates, and that probability of disease transmission is 100% if an uninfected person mates with an uninfected partner.

(Needless to say, the virtue of this model is not realism. Feel free to make your own more realistic example.)

So at time T0, we have .91% of the population infected.

Let p=.1 (i.e. 10% of prudes mate per period).

Then at subsequent periods we have the following expected infection rates:

T1 = 1.77%
T2 = 3.04%
T3 = 4.83%
T5 = 7.27%

If we change p to .2, however (making prudes twice as likely to mate), we get

T1 = 1.79%
T2 = 3.03%
T3 = 4.76%
T4 = 7.12%

This effect doesn't last forever, as there's a later time when the infection rate in the p=.2 world again surpasses that of the p=.1 world. But still, this is pretty counterintuitive.

MJ | April 25, 2007, 1:40am | #

Is Landsburg's essay serious, or is it a variation on "A Modest Proposal"?

"If he takes home an uninfected woman, great -- he distracted her from a potential disease carrier. If he gets herpes, that's also great, because he's sexually conservative and won't pass the infection along very often. Better him than someone with less self control.

Either way, society benefits when the chaste open up slightly."

What individual is going to make a decision about having casual sex based on that criteria?
If the reason he is not interested in one-night stands is fear of STDs, then that does nothing to persuade him, as it increases his personal risk of STDs. If he's a "prude" because he's avoidng making a woman pregnant or simply saving sex for a committed relationship, then what is good for "society" matters not to his decision making. Nor should it, as only if a large number of "prudes" inexplicably change their behavior at one time can this strategy have any effect.

I sincerely hope Landsburg was making a point about misapplying economic theory and the silly conclusions that doing so can create.

jadagul | April 25, 2007, 2:25am | #

MJ: once again, the argument isn't "Having more sex benefits society, so people individually will benefit if they go out and have sex," or even "more people having sex benefits society, so people are obligated to do it." It's "prudes having more sex is a benefit to society, and benefits society more than it benefits the prudes. So we ought to stop taking active steps to discourage them from having sex, and perhaps engage in subsidies (perhaps in the form of free condoms) to encourage them to have more sex."

Or, in other words, the standard argument against free-condom distribution is that it encourages people to have sex more often. Landsburg turns that around and suggests that actually might just be another benefit.

Ken Shultz | April 25, 2007, 2:43am | #

I'm sorry, I still see this as further evidence that the opposite ends in our society strive to become the caricatures their opponents paint them to be.

Not so long ago, I'd have scoffed at anyone who said that the religious right were idealistically like the Taliban, ready to put Jennifer in a burka...

...and now, when I talk to my religious friends about libertarians, I'll have to pretend that none of us really believe that the world would be a better place if only more Christians had herpes.

MJ | April 25, 2007, 7:59am | #

"...the standard argument against free-condom distribution is that it encourages people to have sex more often."

Which presumes the reason "prudes" are not having more casual sex is a fear of STDs which can be solved by increased access to condoms, not that they find little value in casual sex.

Zoomie Junior | April 25, 2007, 8:44am | #

There should be some sort of online profile database of willing, available, and clean potential partners for the sexually underserved... kinda like adultfriendfinder, but with real people.

mobile | April 26, 2007, 6:50pm | #

Has anyone invoked Coase's Theorem yet. If prudes having more sex transfers risk from sluts to prudes but improves the overall welfare, then the logical conclusion is for sluts to pay prudes for sex.