“Since economic models are trained to match climate models, if climate models overstate the effect of CO2 emissions, economic models will overstate the social damages associated with them. “
The fact that CO2 emissions lead to changes in the atmospheric carbon concentration is not controversial. Nor is the fact that CO2 and other greenhouse gases (GHGs) absorb infrared energy in the atmosphere and contribute to the overall greenhouse effect.
Increases in CO2 levels are therefore expected to lead to atmospheric warming, and this is the basis for the current push to enact policies to reduce GHG emissions. For more than 25 years, climate models have reported a wide span of estimates of the sensitivity of the climate to CO2 emissions, ranging from relatively benign to potentially catastrophic, reflecting a wide range of assumptions about how the climate system may or may not amplify the effects of GHG emissions. These continuing uncertainties have direct policy implications.
Economic models for analysing climate policy are calibrated using climate models, not climate data. In a low-sensitivity model, GHG emissions lead only to minor changes in temperature, so the socioeconomic costs associated with the emissions are minimal. In a high-sensitivity model, large temperature changes would occur, so marginal economic damages of CO2 emissions are larger.
While it is common these days for politicians, journalists, and other observers to say the climate is warming “faster than expected,” the data show that, over the past two decades, warming has actually slowed down to a pace well below most model projections.
Depending on the data set used, there has been no statistically significant temperature change for the past 15 to 20 years. Yet atmospheric GHG levels have increased rapidly over this interval, and there is now a widening discrepancy between most climate model projections and observed temperatures.
While a pause in warming is not itself inconsistent with a continuing long term trend, there is no precedent for such a large and continuing gap between models and observations. Some climatologists have argued that within another few years at most, if the pause continues, it will lead inescapably to the conclusion that climate models are oversensitive to GHGs.
Since economic models are trained to match climate models, if climate models overstate the effect of CO2 emissions, economic models will overstate the social damages associated with them. In fact, economic models of climate policy allow for even more exaggerated effects of carbon dioxide emissions than do climate models. Consequently, there is good reason to suppose that economic models too may be subject to revision over the next few years.
One implication of these points is that, since climate policies operate over such a long time frame, during which it is virtually certain that important new information will emerge, it is essential to build into the policy framework clear feedback mechanisms that connect new data about climate sensitivity to the stringency of the emissions control policy.
A second implication is that, since important new information about climate sensitivity is expected within a few years, there is value to waiting for this information before making any irreversible climate policy commitments, in order to avoid making costly decisions that are revealed a short time later to have been unnecessary.
This post is the executive summary of my just released study for the Fraser Institute under the same title. My response to a rebuttal to my report by the Broadbent Institute (Guy at Fraser Institute is pretty sure he just debunked climate change science) was published in the Financial Times, Guy at Broadbent Institute is pretty sure he rebutted report he hasn’t read.