Parts I, II, and III thus far have presented the theory, methods, and policy examples that have historically defined the field of environmental economics. The order of topics has largely followed the intuitive division of the field into the subfields of environmental policy design and non-market valuation. While this organizational scheme was pedagogically useful, it left several important areas heretofore uncovered. Many of these topics fall roughly under the rubric of the “doing” of environmental economics, and so in Part IV we begin by discussing a handful of loosely related topics organized around the theme of cost-benefit analysis (CBA), and then close the book with a wide-ranging discussion of current practice in the field.

Cost-benefit analysis is in some sense the quintessential applied task for environmental and other economists. However, there are important theoretical aspects that need to be understood – particularly when the task involves comparing costs and benefits over long periods of time. We therefore begin in Chapter 21 with a detailed discussion of discounting. We note the common practice of using the market interest rate to compare costs and benefits across time, and present the conceptual basis for this choice, based on the classic result from Ramsey (1928). We then discuss different viewpoints on discounting the distant future, when there are market failures and/or uncertainty about future interest rates. We note how this is particularly relevant for CBA of climate change policies, where the choice of a discount rate is of first order importance.

Climate change CBA is also the primary motivation for our second topic in Chapter 21: integrated assessment models (IAMs). An IAM in general is any modeling system that combines compatible modules from multiple disciplines for the purposes of conducting policy simulations. IAMs are best known for their role in climate change policy, whereby a model of the world economy is paired to a climate system model, and connections between the climate and economic systems are calibrated. As an example of this type of research, we present the analytical aspects of William Nordhaus's DICE model, and show how his stylized representation allows experimentation with different policy scenarios. A general lesson from this is that IAMs are not truth machines, but they are useful for comparing how different policies interact with the economic and climate systems to produce different future outcomes.