Home Page (MVA) > Integrated Assessment Models (IAMs) and Resources > IAMs Thematic Guide
Thematic Guide to Integrated Assessment Modeling
[HOME] [PREVIOUS] [NEXT] [BOTTOM]
Damages Archetype
There are a variety of integrated assessment models, exhibiting a variety of structures. However, many of the different models share a number of core approaches, assumptions, or component models. These shared elements are usually based on earlier models or pioneering work, and might be referred to as ``archetypes''. These archetype models come with their own set of judgments and assumptions, and can have a large bearing on the outcome of integrated assessment modeling studies. It is thus critical that the archetypes be subject to critical scrutiny or review from within the integrated assessment community. Unfortunately, there are already signs that the complete opposite is taking place - wholesale uncritical adoption of archetype models. An example is the link between climate change and GDP losses which is based largely on the work by Nordhaus (1991).
Nordhaus performed pioneering work on estimating the potential impacts of doubled CO2 climate change on the U.S. economy (subsequently extrapolated to the world), and derived an upper bound for damages of about 1 to 2% of GDP. This research is considered to have laid the foundation upon which most subsequent IA impact analyses have relied (Rotmans et al., 1995). As with much of the work on assessing climate impacts, the work was more of a subjective judgment on the part of an expert economist than a true reflection of all the knowns and unknowns regarding climate change impacts. As noted by Funtowicz and Ravetz (1994) the 1-2% GDP loss figure was arrived at in a subjective manner through a combination of quantitative estimates, assumptions, and hunches.
What is disturbing about this is the fact that the 1-2% loss figures have become enshrined in the policy analysis/IA community as being somehow ``reasonable''. Nordhaus' own subsequent work shows that there is high degree of disagreement among experts regarding the potential impacts of climate change. Nonetheless the loss figures persist. In effect, if not by design, the IA community is anchoring to the analytical convenience, and the illusion of understanding that the numbers provide.
There are two lessons to be learnt from this. The first lesson is that experts in the field should provide all the appropriate caveats to the numbers that they put out. To be sure, there is still a problem that work providing all the proper caveats may be taken out of context anyway. However, this does not excuse the analyst from providing the caveats. Perhaps we need to explore whether there are ways of presenting our results that at least minimize such occurrences. There is a second and more important lesson to be learnt by the IA community at large. Uncritical acceptance of the work of disciplinary experts can prove detrimental in the long run, although it may satisfy near term analytical convenience. IA is a complex problem on which little progress can be made without constantly challenging new and accepted ideas, even if these ideas come from within the disciplinary communities to which many IA researchers belong.
