Environmental 5: IAMs Flashcards
Why do Stern and Nordhaus disagree over the optimal scale of climate policy?
Nordhaus and Stern disagree on the choice of discount rate. Nordhaus sets the pure rate of time preference rho equal to 1.5%, and the elasticity of the marginal utility of consumption eta equal to 1.45. This generates a discount rate of 4%, by the Ramsey discount formula. Stern, by contrast, argues that there are no ethical grounds to claim rho>0, except a small catastrophe risk epsilon=0.1%. Combined with a lower elasticity of marginal utility of consumption eta=1, he uses a discount rate of 1.4%, which is significantly lower than Nordhaus. Since the major costs of climate change fall in the far future, higher discounting shrinks these costs considerably in present value calculations, and the marginal cost of abatement equals marginal damages at much lower levels of abatement.
What is ‘fat-tailed risk’? What conclusion does it lead Weitzman to over optimal climate policy?
Even if the tails of a probability distribution decline to zero, they may do so slowly enough that the distribution does not have a finite expectation. Examples include the Pareto or log-logistic distributions. Weitzman claims that the pdf of extremely bad outcomes from climate change is fat tailed, and that the welfare consequences of bad climate outcomes grow rapidly with temperature. His ‘dismal theorem’ therefore concludes that the social cost of carbon is very sensitive to the worst possible outcome. We should therefore see action on climate change as insurance against very extreme, very unlikely events.
Why think that climate change might affect capital stocks, as well as output? What effect does this have on optimal policy?
Climate change might damage capital directly, through extreme weather events or abandoned coastal areas. Some assets will be ‘stranded’ as they were designed for a world with a different climate regime, reducing the productivity of capital as some will be needed to restore and repurpose old stocks. Introducing this effect to the DICE model increases the social cost of carbon, sharply so when the capital damage function is more convex than quadratic.
What is the setup of the Acemoglu et al (2012) model?
Output is produced using a CES production function between clean and dirty output. Output in each sector changes over time with the number of scientists working in each area. Outputs concentrate where the technology is best, creating the highest profits in the most productive and producing sectors. Scientists will work where profits are highest. Unless a subsidy attracts them to the other sector, this will be the sector with the best technology.
When, in the Acemoglu model, is a temporary subsidy to clean research enough to prevent an environmental disaster?
If the two inputs are strong substitutes, a temporary subsidy and a temporary Pigouvian tax can incentivise scientists to begin working in the clean sector. Soon, technology will be good enough in this sector that the policies are no longer needed, as the clean sector is more productive and profitable anyway. An environmental disaster is prevented as the dirty sector withers away.
When, in the Acemoglu model, is a permanent subsidy to clean research, plus a permanent Pigouvian tax, enough to prevent an environmental disaster?
If the two inputs are weak substitutes, the temporary subsidy cannot prevent the environmental disaster. A permanent Pigouvian tax is needed to continually incentivise clean output and avoid the disaster.
If the two sectors are complementary, the only way to avoid environmental disaster is to stop long-term growth.
Why do Sterner and Persson (2008) think that strict climate policy is justified on the grounds of imperfect elasticity of substitution between environmental goods and other consumption?
Consumption and the environment are imperfect substitutes. The marginal value of environmental services is not constant with E/C, as assumed by single-good IAMs. Environmental services become more valuable the more scarce they are. Therefore, the marginal value of C decreases as E is damaged. In an IAM, they find that this effect justifies even stricter climate policy than Stern, due to the extreme environmental costs of Nordhaus-policy.
