Ezra Klein has weighed in on the global warming with a couple of posts (h/t Andrew Sullivan). In the first he says of the back-and-forth here between Bradford Plumer and me:

But they both ignore a point that's central to Manzi's argument: What happens after 100 years?
Letting greenhouse gases build in the atmosphere is a bit like letting a tree grow roots beneath the foundation of your house. It may not be that bad this year, or next year, or even the year after that. But with each year that goes by, the problem becomes incrementally more severe, and harder to reverse. So even if Manzi is right that the costs are manageable into 2100 -- a century, after all, is a long time for a human, but not for the atmosphere -- what does that do to our descendants who have to deal with a scorching planet between 2100 and 2200? And then into 2300, and then 2400?

When thinking about taking actions now to shape the future environment, we should start with the recognition that our ability to make meaningful predictions generally declines as we look further and further into the future. This proceeds in shades of gray from, illustratively, “2030” to “10,000AD.”

At several points in my post I described the projected impacts of climate change through about the year 2100. This is because numerous IPCC forecasts are done through about that point in time, due to the view that projections beyond this point are too speculative.

When thinking about time after 2100, we have, in cartoon terms, two choices: (i) simply treat it as unknowable fog, or (ii) attempt to guess. I think that if we take the first choice, then we simply try to forecast through the next century, and let future generations worry about the world beyond 2100 (though I’ll point out that it is a very unusual political debate in which we call trying to manage the entire world for about the next 100 years as “short-termism”).

If we take the second approach, how far out do we try to guess? The Nordhaus economic calculations that I cited in my post as formal attempts to compare odds-adjusted costs versus benefits actually extend out for 250 years. That is, they consider expected costs and benefits to about 2250. Therefore, Klein’s point is really about potential damages beyond 2250,not 2100. That’s a long way off.

This doesn’t mean that I don’t care about problems that might occur hundreds of years from now, just that I don’t care much about current predictions about those problems.

Advanced economies are on a very long-term path of de-carbonization, driven far more by technology development than anything else. We don’t know whether or not this will continue, or what the technology, social, or political landscape of the year 2250 will look like; but under any IPCC scenario that calls for growing emissions, it would be dramatically wealthier and more technologically advanced than our own. For all I know, we will have materially de-carbonized by then in the absence of emissions restrictions because of technology breakthroughs that are as speculative to us today as nuclear power, space travel, and the Internet would have been in 1750. Or, we might engage in a series of nuclear and bio-weapon global wars in the 22nd century that makes CO2 emissions a trivial problem. Or, we might develop good scrubbing technology 127 years from now that could solve the problem for 1/1000th of the cost of restricting carbon emissions today. And so on. In my opinion, it is hubris for us to imagine that we should be turning dials on our emissions now, because of the effects we project this will cause sometime past 2250.

In Klein’s second post, he restates the Precautionary Principle as follows:

There's a range of likely outcomes from a tax on carbon, and we can handle most of them. There's also a range of outcomes from radical changes in the planet's climate, and we've really no idea which we can handle, and which we can't.

This sounds wise and reasonable on the surface—but I’ll refer readers back to my original post (or Cass Sunstein’s book) to explain why I am highly confident that it is a useless principle for making actual decisions. The shortest form version of this is simply to note that I can substitute a thousand different pairs of text strings into these sentences in place of “a tax on carbon” and “radical changes in the earth’s climate”, and the identical logic would apply. Try “investing 25% of global GDP every year into asteroid interdiction” and “having an asteroid as big as the one that killed off all the dinosaurs smash into the earth”; or, “not permitting homosexuals to marry” and “radical changes in our social order”; or, “investing 50% of global GDP every year into attempts to ameliorate climate change” and “radical changes in the earth’s climate”, etc. In effect, it is an argument against ever making any change, and also against ever choosing to forgo any insurance policy against any risk, as long as the cost of the insurance is at a price that humanity “can handle.”

No, you say, that’s ridiculous—the Precautionary Principle is an argument only against what a reasonable and informed person would believe to be a legitimate danger. There’s the rub. The hard work in the decision logic is to attempt to handicap the risks of some potential action, as well as of the various alternative courses of action to avoid or ameliorate these risks—recognizing that total resources are finite, and that how we address any one of these challenges will affect our options for addressing all of the others. The world has invested an enormous amount of money into exactly such research for the challenge of climate change, as collated and reviewed by the IPCC process. I advocate using this information to inform our decisions.