“Passing out condoms increases the amount of sex but makes each sex act less dangerous. So theoretically it’s indeterminant whether it increases or decreases the spread of AIDS.”
Not quite—on a rational choice model, passing out condoms may decrease or not impact the spread of AIDS (in principle), but it can’t increase it. A rational actor who doesn’t actively want AIDS might increase their sexual activity enough to compensate for the added safety of the condom, but they would not go further than that.
(This is different from the seatbelt case because car crashes result in costs, say to pedestrians who are struck, that are not internalized by the driver.)
We might suppose that condom promotion has two effects: a replacement effect and an encouragement effect. So there will be instances where what would have been unsafe sex at all becomes safe sex, and some instances where no sex at all becomes safe sex. Safe sex is so vastly less likely to transmit HIV that the latter effect would have to be hundreds of times larger than the former for condom promotion to have an overall increasing effect on HIV transmission; that doesn’t seem plausible to me and no evidence to support it has been presented.
If you could show that condom promotion caused a lot of instances where no sex becomes unsafe sex that would change the picture, but AFAIK there’s no reason or evidence to suppose that.
It’s pretty clear in this instance that the desire to bash the Pope-criticising liberals came first, and the arguments second.
In theory—I say nothing of practice—this need not be true. If people get ten times as much sexual pleasure per unit risk, they may pay out more total risk. As a general principle of resource consumption this has an official name, but I forget it.
Obviously we can construct an agent who does this. I just don’t see a reasonably parsimonious model that does it without including a preference for getting AIDS, or something similarly crazy. Perhaps I’m just stuck.
Are you talking about the idea of a Giffen good, where bread becomes more costly, reducing your household budget, forcing you to get more of your calories from cheaper goods such as bread, meaning you buy more bread? If so I’m not sure I see how that applies in this instance.
In general I don’t mean to make an argument that it’s impossible for harm reduction strategies to be counterproductive, but that condoms in particular are so very effective at reducing HIV transmission that it’s simply implausible to posit that any countervailing effect could do so much as to make the overall outcome worse—such an idea would need very strong evidence, which plainly isn’t here.
“Passing out condoms increases the amount of sex but makes each sex act less dangerous. So theoretically it’s indeterminant whether it increases or decreases the spread of AIDS.”
Not quite—on a rational choice model, passing out condoms may decrease or not impact the spread of AIDS (in principle), but it can’t increase it. A rational actor who doesn’t actively want AIDS might increase their sexual activity enough to compensate for the added safety of the condom, but they would not go further than that.
(This is different from the seatbelt case because car crashes result in costs, say to pedestrians who are struck, that are not internalized by the driver.)
We might suppose that condom promotion has two effects: a replacement effect and an encouragement effect. So there will be instances where what would have been unsafe sex at all becomes safe sex, and some instances where no sex at all becomes safe sex. Safe sex is so vastly less likely to transmit HIV that the latter effect would have to be hundreds of times larger than the former for condom promotion to have an overall increasing effect on HIV transmission; that doesn’t seem plausible to me and no evidence to support it has been presented.
If you could show that condom promotion caused a lot of instances where no sex becomes unsafe sex that would change the picture, but AFAIK there’s no reason or evidence to suppose that.
It’s pretty clear in this instance that the desire to bash the Pope-criticising liberals came first, and the arguments second.
In theory—I say nothing of practice—this need not be true. If people get ten times as much sexual pleasure per unit risk, they may pay out more total risk. As a general principle of resource consumption this has an official name, but I forget it.
Obviously we can construct an agent who does this. I just don’t see a reasonably parsimonious model that does it without including a preference for getting AIDS, or something similarly crazy. Perhaps I’m just stuck.
Are you talking about the idea of a Giffen good, where bread becomes more costly, reducing your household budget, forcing you to get more of your calories from cheaper goods such as bread, meaning you buy more bread? If so I’m not sure I see how that applies in this instance.
In general I don’t mean to make an argument that it’s impossible for harm reduction strategies to be counterproductive, but that condoms in particular are so very effective at reducing HIV transmission that it’s simply implausible to posit that any countervailing effect could do so much as to make the overall outcome worse—such an idea would need very strong evidence, which plainly isn’t here.