I don't think that the discount rate used in this model has anything to do with the resource itself or with any other parameters in the model. It's actually driven by the performance of the rest of the economy and simpy tells you how much your money in the bank will appreciate with time. That's what creates the problem, since it's some sort of artificial index that has little connection with real life, just like money flows become detached from flows of actual goods and services.

That's just part of the problem. If we were operating within some fair economic system driven by only market forces, we should already be seeing some signals and reactions to them. Indeed if the oil prices are growing at 30% while the discount rate is 8-10% max, somebody should already start realizing that keeping more in the ground, is a better deal than pumping it out, converting into $$ and reinvesting. Unfortunately it's not just economic decisions (no matter how bad they can be) that regulate the system. All the markets are distorted by subsidies, politics, private and corporate interests, etc., so the prices as they are are not realistic, and economic models have very limited applicability.

What's nice about Gowdy's paper is that it explains some pretty obvious dynamics (if you have a limited resource and use it faster, then you get less left and it will hit you harder and sooner when you run out of it, duh?) using models and terminology that economists can finally understand. So there is hope.

Well, I'm not sure what the intention is in the Hotelling case, but generally the discount rate used to, say, evaluate an investment, contains a "risk premium" that depends on the uncertainty of expected benefits/paybacks. For example, see the wikipedia definition of "discount rate": http://en.wikipedia.org/wiki/Discount_rate

"Typically, the discount rate is arrived at by beginning with the appropriate interest rate for the length of time in question, then adding an additional sum to account for risk. For example, some companies add 15% to term-specific risk-free rates."