Redefining Integrated Assessment Models

Redefining Integrated Assessment Models: An Exploratory Approach Towards Robust Climate-Economic Policies

Lingeswaran, Shajeeshan (TU Delft Technology, Policy and Management)

Storm, Servaas (mentor)
Kwakkel, Jan (mentor)
van Beers, Cees (mentor)

Delft University of Technology

Engineering and Policy Analysis

2019-08-20

Integrated Assessment Models (IAMs) are aiming to shed light on the cost-benefit of climate mitigations. However, current IAMs are depicted with a wide range of weaknesses. Next, to questionable assumptions of model functions, such as the damage function, IAMs are inadequate at addressing deeply uncertain parameters, such as the equilibrium climate sensitivity (ECS). Various research has been conducted in addressing uncertainties by utilizing stochastic dynamic programming or approximate dynamic processing. However, two crucial aspects were not considered in these studies. First, most researchers utilized optimization to determine the optimal policy as their decision analytic method for the risk analysis. Yet optimal strategies are highly sensitive to uncertainties and thus, they lose their prescriptive value in a deeply uncertain environment like in the field of climate economics. Furthermore, most economists described the deep uncertainties of IAMs model by a normal distribution, at best by a lognormal distribution. However, Weitzman (2009) has shown in his Dismal Theorem, that deep “uncertainty in the form of fat tails is, at least in theory, capable of swamping the outcome of any CBA”. This research addresses both aspects by utilizing the Exploratory Modelling and Analysis (EMA) framework on the Dynamic Integrated Model of Climate and the Economy (DICE) of the Nobel prize winner Nordhaus. By translating the DICE model into a stochastic simulation model and conducting an EMA approach, this research demonstrates the importance of defining robust strategies against the disproportional risk in the fat-tails.

Intgrated Assessment Models
Decision-Making
Deep Uncertainty
Fat-tailed Distributions
Robust Optimisation

http://resolver.tudelft.nl/uuid:812be430-79fb-4900-a0e2-b0bc9c3d804b

Student theses

master thesis

© 2019 Shajeeshan Lingeswaran