« The federal budget: Austerity or stimulus or neither? | Main | The generosity collapse »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Given the practice of the cities of Vancouver and Victoria from dumping sewage into the Strait of Georgia, remind me not to drink water the next time I am in White Rock. The water comes from the mountains of North Shore of Vancouver. Can be special but we have not found a way to live in salt water yet.

Actually, the White Rock water is from a deep well and is, by choice, not treated with chlorine. It looks and tastes much better than the water delivered from the salmon-destroying reservoirs located in North Vancouver. Though I'm sure that Victorian lawns do not care.

I mentioned White Rock's proximity to the Pacific Ocean because I was about to rant on about how the municipality of White Rock uses 2-stroke leaf blowers to clean up sidewalks and similar. You see White Rock enjoys excellent air quality (and much less rain than North and West Vancouvers). But the Fraser Valley airshed occasionally plugs up with smog and a healthy amount of small particulate matter that makes people sick and kills a few others.

I thought the use of unchlorinated water and 2-stroke leaf blowers at the healthy, clean air end of the Fraser Valley airshed made for an interesting constrast.

Boy am I late to this thread. Perhaps it would be better to widely publish the findings, then make lithium available over the counter as a food additive or in pill form. If it's establish that people who consume regular amounts of the element have better mental health, then people will feel more comfortable with the idea of taking lithium.

K-"That's how you model the things you know. But, you've neglected the question of whether the expected value of the things you never imagined, and therefore never considered, is negative. But if we agree on the the correct recourse, maybe we can agree to disagree about the reasons."

So, we should assume that the result will be somewhat worse than what the model predicts. The question is how much. You seem to be advocating that we discount all expected values absolutely when losses relative to the status quo are possible. First of all, I think losses relative to the status quo are a possible outcome of every decision (even the decision to do "nothing"). Second, I don't think absolute discounting is optimal. Maybe it is sometimes, depending on how much we know (or, more accurately, how confident we are in what we know), but I think it's often better to go with a good guess than to cling to what knowledge in which we have absolute confidence (which I think we would agree shouldn't be ANY knowledge).

I would like to add that there are really multiple connected issues being argued here. Some people are arguing about whether lithium is/could be beneficial in the ways that Mike originally suggested. Others, like myself, are interested in the issue of whether such benefits would warrant government action, assuming both that those benefits were observed and that political systems would allow that intervention. Others are trying to address the question of whether it would be beneficial to society to allow these sorts of government interventions in the first place. These are all interesting questions, but I don't feel particularly qualified to make much comment on the first or the last, hence my focus on the second issue. However, I would like to make it clear that while my arguments would be irrelevant if the first or last question were answered in the negative, I'm entirely open to those possibilities. I'm agnostic at this point about them both.

Robert Enders-"If it's establish that people who consume regular amounts of the element have better mental health, then people will feel more comfortable with the idea of taking lithium."

But what if, for the sake of argument, the "mental health" benefits were entirely external to the person consuming the lithium . More generally, the issue is what to do if lithium consumption involves an externality.

I'm not advocating some arbitrary additional discounting. I'm just saying that the model expectation represents a lower bound on the expected harm, thus prudence is advisable. I don't even advocate focusing on the first moment or even the probabilities of disaster. Assume that bad things are going to happen and worry about what we are going to do when they do. And above all, place the burden of proof of safety on those who expect to profit from the project, and find ways to ensure that they will have the means to make restitution when things fail.

The comments to this entry are closed.

Search this site

  • Google

Blog powered by Typepad