This is CBSI in the world. I'm John Boucher with Professor Richard Epstein of the Hoover

Institution. We're discussing a bold initiative by Gary Gensler, the new chair of the SEC,

formerly a chair of the SEC during the Obama administration. He's returned to his post

to present to all companies, private market companies, a requirement to disclose what their

investments mean for the future of the Earth's climate. And the second question that Richard

raises in his most recent essay, Defining Ideas, is, can this proposed initiative, in other

words, disclosure, can it lead to curbing what you presume right now are negative effects

of global warming? How would they demonstrate that, Richard?

Well, I mean, look, when you're starting to do this, it's pretty clear that you know what to

do when you have relatively closed systems. So, for example, I suppose you have a bunch of

city shells and they're at the edge of a river, but they have to have a salty environment in order

to survive. If it turns out that you pour out too much pure water, it may not be a pollutant

in the ordinary sense, but what it's going to do is to dilute the salt so the organisms will die.

And so, therefore, you have a relatively clear recipe of what it is that you have to do

when it comes to controlling those kinds of flows. But when you're dealing with global warming,

or you're dealing with what is called a complex chaotic system, which doesn't mean that it's just

random events. It means that these events are fully determined by the usual principles of causal

science, but you don't know the initial conditions well enough to do it. And the key thing about

the environmental prediction stuff is very small changes in the initial positions can result in

vastly different changes in the consequence. So, we're talking about things which are extremely

crude, you know. We're lucky to get this stuff with an order of magnitude. When you're trying to

do the predictive side of this, if you change something by a half a degree, it could result in

...