More simulations on carbon tax, GHGs and economic impacts
In my post yesterday, I lamented the fact that the Jaccard modeling of carbon tax impacts for the federal government assumed uniform carbon tax rates applied immediately and held constant over time. But in the real world, some phase in period is going to be necessary.
Enter the Interim Report of the National Round Table on the Environment and the Economy, and a new round of modeling carbon taxes. It should be noted that these simulations have a common pedigree with those of Jaccard. The report’s modeling was done with the same CIMS simulation package, via J&C Nyboer and Associates. It turns out that:
Nyboer is the executive director of the , part of Jaccardâ€™s research group at SFU. Nyboer uses a computer tool called the Canadian Integrated Modelling System (CIMS), a product of Jaccardâ€™s PhD thesis. CIEEDAC has standing contracts with Statistics Canada, Natural Resources Canada, and Environment Canada to evaluate energy use by Canadian industry.
At some point, we will have to look under the hood of CIMS to see what makes it tick. Statistical models generally use data and empirical evidence, plus lots of guesswork, to get their model to approximate a base year, then allow for tweaking of the parameters in order to conduct the simulation exercises. For example, we do not really know what the price elasticity of gasoline is when we contemplate a doubling or tripling of prices. It is very important to understand what is driving the bus here, but for the moment, I will not make a methodological critique.
The new study sets out four scenarios of a phased-in carbon tax: two each for reductions of 45% and 65% below 2003 GHG levels, with both a slow start (slow rise in carbon tax but higher end point) and a fast start (quicker rise in carbon tax then flattening out after 2035). Most of the interesting stuff is on pages 10-12 (they also do some modeling about air pollution, but I am more interested in greenhouse gases).
What is interesting about the results is the magnitude of tax required to meet the targets. All scenarios start modestly with a $10 per tonne of CO2 tax in 2010 rising to $15 per tonne in 2015; thereafter they scale up at different rates. Thus, by 2015 not much happens in terms of CO2 emissions. The real action hits when the carbon tax gets higher, and they conclude (p. 2):
To attain a 45% reduction target, the price is estimated at $160 and $200/tCO2e in 2050. To attain a 65% reduction target, price estimates are $270 and $350/ tCO2e in 2050.
Interestingly, the economic costs are not particularly high with the most aggressive scenario coming up with a GDP in 2035 that is 2.2% less than business-as-usual (the impact then mostly peters out by 2050). The report summarizes the economic impacts (p. 10):
[T]he most adjustment occurs from 2030 to 2045, with the economy re-stabilizing on its new less GHG intense path by 2050. While the annual reduction in GDP for the four scenarios lies between 0.90% and 1.44% of BAU [business-as-usual] GDP per year, when averaged over the forty year period, the individual yearly costs over the period varying from 0.1% to 2.3% per year (Table 4). The reader should note these costs are NOT cumulative â€“ they are subtracted from potential GDP in the year in question.
If anything, based on the Jaccard research, we should examine further scenarios that implemented a green tax at higher rates sooner. It may well be the case that by 2050 we need something more like an 80% reduction in CO2 emissions. And finally, the report says we need to get busy:
Regardless of the medium-term target chosen, if emissions reductions are pushed into future periods, the overall costs of achieving the reductions will rise. A later start implies that the GHG price signal eventually has to reach higher levels to produce the same level of GHG reductions, and that a very late start will jeopardize reaching any deep reduction target at all.