In a provocative post, Joe Romm argues that even “strong climate action” would be well worth the effort. Yet Romm’s claim that stabilizing atmospheric greenhouse gases at 445–535 ppm (CO2-eq) would cost only “one tenth of a penny on the dollar” (through 2050) understates the IPCC’s actual cost estimate by about 95%. In reality, the IPCC’s reported estimate translates into a mitigation cost of about 2.2 cents on the dollar–far far higher than Romm’s figure. Romm’s mistake has nothing to do with climate science: he simply confuses the rate of growth in income, with income itself.
To make matters worse, even when correctly interpreted, the IPCC estimate significantly understates what the cost will be in practice. The IPCC admits that its estimate is a theoretical textbook case, which assumes all participating countries implement their mitigation policies perfectly, and keep them in force throughout the 21st century. To the extent that any major government deviates from this flawless enforcement, even if just for a few years during a severe recession in (say) the years 2030–2032, then the costs of achieving the stipulated atmospheric concentrations will rise above the stated figure.
Presumably none of these details will alter Joe Romm’s conclusion. After all, he contrasts the slower economic growth resulting from new government regulations to the “incalculable cost of catastrophic global warming impacts to the next 50 generations.” With such a specter in mind, it hardly matters how much potential income the world will forfeit in order to meet Romm’s desired concentration targets. Still, for those who do think magnitudes are important, I explain below exactly how Romm managed to so severely understate the IPCC’s own estimate of the costs of mitigation.
Romm Unwittingly Counts Only the Costs in Year 1. Romm takes the IPCC’s figure of mitigation cost and attempts to translate it into common language, but in the process renders it meaningless. The IPCC’s Fourth Assessment Report (AR4) Table SPM.6 [pdf] estimates that stabilizing the atmosphere in the range of 445–535 ppm (CO2-eq) would reduce global GDP in the year 2050 by (at most) 5.5% from the baseline trajectory. The IPCC table also has a column showing that this figure works out to a slowing of GDP growth by less than 0.12% per year. This is why Romm says, “So global GDP drops by under 0.12% per year — about one tenth of a penny on the dollar.” This is Romm’s main point of the post; he even works this into the title. But it is meaningless; the IPCC figure of 0.12% per year doesn’t imply what Romm says that it does. Quite simply, Romm is taking their figure of a slowdown in the rate of growth of income, and interpreting it as a statement about the absolute drop in income.
If we just focus on the year 2050, humans will lose (up to) 5.5 pennies on the dollar. That is equivalent to saying that a household earning $100,000 per year in the baseline scenario would pay $5,500 in the year 2050 alone, because of the government’s constraint on GHG emissions.
Now it’s not correct to conclude that this stabilization target would cost “5.5 cents on the dollar,” because that figure only applies to the year 2050. In prior years, the gap between business-as-usual GDP and CO2-stabilized-GDP grows smaller, the earlier back we go in time. That’s where the 0.12% figure comes from, after all: The government-constrained GDP grows a little more slowly than the baseline GDP, such that every year the former falls further behind. By the year 2050, that slight disadvantage–a slower annual growth rate of 0.12%–has grown exponentially into an accumulated 5.5% deficit.
Thus, we see that Joe Romm’s claim of “one-tenth of a penny on the dollar” is only true for the first year of the program’s implementation (assuming the constrained GDP grows at a constant 0.12% lower rate than unconstrained GDP). By the second year, the constrained GDP would have fallen to 0.24% (plus some change because of compounding) behind the unconstrained, business-as-usual GDP. In that second year then, the globe would be sacrificing more than “two-tenths of a penny on the dollar” because of the GHG caps. And to repeat, by the year 2050, the globe would be paying “5.5 pennies on the dollar.” So Romm’s statement completely miscontrues the economic significance of the IPCC’s figure of an average slowdown of 0.12% in annual GDP growth.
If we wanted to come up with a single number to gauge the overall impact on income (or what is the same thing, real GDP), we would need more information than what is provided by the IPCC summary tables. Technically we are trying to say what the percentage reduction would be in the present discounted value of global income through some future date, say 2050, because of government caps on GHG emissions. In order to do that properly, we would need to know (a) the precise GDP losses (from mitigation efforts) in each year, and (b) the appropriate discount rate to use. The IPCC summary does not provide such detail; it only tells us the GDP loss in year 2050.
But let’s give a ballpark illustration: Suppose the stabilization path means that GDP grows 0.13% more slowly than in the baseline scenario, starting in the year 2009 and running through 2050. This means GDP will be 5.47% lower in 2050, so our numbers are calibrated with the IPCC’s projection. (We are using 0.13% > 0.12% because there is less time now than when the AR4 came out; we need to have constrained GDP lag farther behind unconstrained GDP, since we have fewer years to work with than when the AR4 numbers were computed.) So for a household that would normally make $100,000 per year in the baseline, the mitigation policies cost $130 in 2009, $260.17 in 2010, $390.51 in 2011, …, $5334.02 in 2049, and $5470.95 in the year 2050.
If we further assume a 3% interest rate, then the present discounted value of the 41-year stream of mitigation costs is about 2.2% of the PDV of the 41-year stream of $100,000 earnings. (Fiddling with the interest rate would move the cost percentage in the opposite direction, though not by much. For example, if we use a 1% discount on future income, then the IPCC’s 2050 cost estimate translates into 2.6% in lost GDP over the 41 years, while a 5% discount rate implies a loss of just 1.9% of total output over the period.) Thus Romm’s claim that stabilizing at 445 ppm of CO2-eq would cost “one tenth of one penny on the dollar” is off by a tad. A much more accurate figure–and one completely calibrated with the same IPCC estimate Romm uses–would be 2.2 pennies on the dollar.
In short, Romm misinterprets the IPCC figure, thereby underestimating the blow to income from mitigation policies by about 95% of the IPCC’s high-end estimated value (assuming we discount future income at 3%). To avoid confusion, let me reiterate that in order to compress a stream of yearly income losses into a single number, we need to pick some discount (interest) rate. We have here (arbitrarily) picked 3%, and Romm or others can quibble with that. Yet they should be cautious: If they argue for a lower discount rate, that will inflate the present discounted value of the costs of mitigating GHG emissions. In any event, Romm’s figure is completely out of bounds–reflecting the fact that he misinterpreted what the IPCC number of 0.12% meant.
The IPCC Estimate Itself Relies on Textbook Enforcement for a Century In the section above we demonstrated that Romm miscontrued the IPCC’s estimate of (up to) a 5.5% reduction in world GDP by 2050 from aggressive stabilization policies. However, even that estimate itself is a best-case scenario. As the AR4 WGIII report tells us on page 204:
It is important to note that for the following reported cost estimates, the vast majority of the models assume transparent markets, no transaction costs, and thus perfect implementation of policy measures throughout the 21st century, leading to the universal adoption of cost-effective mitigations measures, such as carbon taxes or universal cap and trade programmes….Relaxation of these modelling assumptions, alone or in combination (e.g. mitigation-only in Annex I countries, no emissions trading, or CO2-only mitigation), will lead to an appreciable increase in all cost categories.
Conclusion Climate change is not a subject to be taken lightly. Some very credentialed scientists have generated computer simulations in which unchecked GHG emissions eventually impose significant damages on our descendants. Be that as it may, it is necessary to come up with accurate assessments of the likely economic damages we will impose on these same descendants in the form of government mitigation policies. Joe Romm’s treatment understates the IPCC (high-end) estimate by about 95% through simple confusion about their number, while that number itself relies on extremely implausible assumptions about government behavior.
In his post, Romm went on to make numerous claims about the possible economic benefits from capping GHG emissions. In a future post, I will address these claims. For the present post, I wanted to focus on the problems with his “one-tenth of a penny on the dollar” claim.