Easy, Cheap ‘Green’ Energy? Just the Reverse!

By Kenneth P. Green -- February 26, 2010

[Editor note: This post by Kenneth P. Green and Aparna Mathur of the American Enterprise Institute, is a slightly revised version that originally appeared at The American, AEI’s flagship monthly publication.]

In December 2009, economists Hector Pollitt and Chris Thoung of Cambridge Econometrics published a self-described “short” modeling exercise on an 80 percent greenhouse-gas emissions reduction by 2050 in the United Kingdom.[1] Pollitt and Thoung used the Energy-Environment-Economy Model of Europe (E3ME), which they observe has been “used for a variety of analyses including greenhouse-gas mitigation policies, incentives for industrial energy efficiency, and sustainable household consumption.”[2] The E3ME model covers 29 European countries and uses detailed data on 42 economic sectors, 41 categories of consumer goods, 12 types of fuel, and 14 emissions, including the six major greenhouse gases.

Surprisingly, the study estimates that prices for most goods would rise by less than 1 percent. The highest price increases would be found for carbon-intensive fuels, such as natural gas, gasoline, and electricity. But it would be affordable overall.

Proponents of (forced) energy transformation trumpeted the findings. In a news piece “Low-carbon future: we can afford to go green,” Jim Giles of New Scientist quoted a climate policy expert at the London School of Economics: “These results show that the global project to fight climate change is doable” and “it’s not such a big ask as people are making out.”[3]

The New Scientist piece adds that these results correspond well to studies in the United States, quoting Manik Roy of the Pew Center on Global Climate Change as predicting that “even cutting emissions by 80 per cent over four decades has a very small effect on consumers in most areas.” The Pew Center would no doubt like to see a lot of rationing of CO2 here in the U.S.

No Inexpensive ‘Green’ Lunch

We disagree. Our prior research into the costs of indirect energy—that is, the energy used to produce, package, and distribute consumer goods and services—suggests that the Cambridge Econometrics numbers are unrealistically low. In a series of papers and studies conducted by us and coauthored with colleagues, we show that even a $15 permit price (one-fortieth of what Pollitt and Thoung model) would cause prices of most goods to rise by 1 percent or higher.[4] So hold on to your wallets.

Assumptions and models, naturally, explain the differences between our estimate than theirs. Let’s consider the major ones.

Pollitt and Thoung compared two “projections” with the E3ME model: a business as usual baseline projection extrapolated to the year 2050 and another scenario which assumes that the United Kingdom’s 80 percent reduction target is met by 2050. The two mechanisms modeled to attain the 2050 target were the European Union Emissions Trading System (EU ETS), with a cap that declines 3 percent per year, and a carbon tax levied on the sectors of the economy that do not trade in the EU ETS. Pollitt and Thoung also had to assume some significant technological achievements, including a switch from natural gas to electricity in the domestic sector, and an impressive (if implausible) assumption that electric vehicles will reach a market penetration of 90 percent by 2050.

Projecting forward to 2050, the E3ME model indicates that to achieve such reductions, the carbon price or the price of carbon permits would rise to almost €410 (about $600) per ton of CO2[j2] for the traded sector, and €210 (about $300) for the non-traded sector. These high prices are then passed forward to consumers in the form of higher prices for consumer goods.

Our own model, by contrast, uses input-output matrices to examine the effect of a carbon price increase on everyday consumption items. The direct impact of a carbon price is on fossil fuel costs, while the indirect effect comes from the use of these fuels in the production of consumption items. Our analysis, described in detail in Hassett, Mathur, and Metcalf (2009), was conducted for a period of three years: 1987, 1997, and 2003. We can use the results from this paper to see how a much higher price increase would translate to higher consumer goods prices.

To do so, we simply scale up our derived percentage price increases by the ratio of the estimated carbon price under the E3ME model to our assumed carbon price of $15 (all prices are in 2009 U.S. dollars). The table below shows the price increases that our model would predict if carbon prices did, in fact, rise to the level predicted by the E3ME model. As is amply clear, the price increases are likely to be significantly higher than 1 percent. The biggest price increases would be seen in fuel costs (to the tune of nearly 500 percent for electricity and natural gas) but would also be significant in food costs (20 to 30 percent) and transportation. So what, exactly, is causing the divergence in results between the Cambridge Econometrics numbers and our estimates?

Projected Price Increases (percent)

Food at home	27.40
Food eaten out	22.70
Food at work	33.66
Tobacco	26.22
Alcohol	22.70
Night club	22.70
Clothes	15.66
Jewelry	16.83
Toiletry	28.18
Health/beauty	16.44
Home rental	12.13
Other rentals	16.44
Furnishings	21.53
Household supplies	27.79
Electricity	491.22
Natural gas	480.66
Water	24.66
Home fuel	374.19
Telephone	10.18
Health	15.27
Autos	35.23
Auto parts	25.44
Gasoline	302.56
Car insurance	12.13
Mass transit	35.23
Other transit	24.27
Airfare	72.80
Books	13.31
Pubs	19.18
Recreational sports	16.44
Other recreation	19.96
Gambling	12.13
Higher education	11.74
Lower education	13.31
Other education	11.74

Why the Divergence?

There are several explanations for the sharp difference between our findings and those of Pollitt and Throug.

Our data relates only to the U.S. economy, while their results are derived from a simulation of the EU economy. Our analysis is not dynamic, in the sense that it does not allow for consumers and firms to alter their behavior in the face of rising prices, as their study does. Their study claims that the government revenues from the tax are used to encourage people to switch from natural gas to electricity in the domestic sector and to use more electric vehicles. The exact modeling of this is unclear. Also, prices in their model are affected by prices in other countries since certain products are tradable across regions.

However, even given these considerations, the Cambridge Econometrics results seem highly implausible. For instance, the assumption that by 2050, the market penetration of electric vehicles will be almost 90 percent is unrealistic given current and historical rates of adoption. There are currently no mass-market electric vehicles on the market. In Europe, fleet turnover involving only a change in fuel type (rather than a change in the nature of fuel) can take several decades.[5] Also, the assumption that the domestic sector will easily and quickly adapt and move from natural gas to electricity is questionable. A more realistic scenario is one which lies closer to our estimates, so that the dynamic response does lead to a lower incidence of the carbon tax over time, but not as much as assumed under the Cambridge Econometrics model.

Conclusion

Supporters of greenhouse gas controls often portray them as cheap and easy. But they must resort to economic models that rely on dubious assumptions of future rates of technology development and market penetration. A more straightforward input-output models suggest greenhouse gas controls will be anything but cheap, and they certainly won’t be easy. However, even small steps toward carbon rationing are proving to be unaffordable and politically dubious.

Reality bats last in the energy debate.

[1] Hector Pollitt and Chris Thoung (2009). “Modeling a UK 80% Greenhouse Gas Emission Reduction by 2050, A short modeling exercise for New Scientist” Cambridge Econometrics.

http://www.newscientist.com/data/doc/article/mg20427373.400/ce_new_scientist_report.pdf

[2] Ibid, p. 1.

[3] Jim Giles, “Low-carbon future: we can afford to go green,” New Scientist, December 2, 2009. Available at http://www.newscientist.com/article/mg20427373.400-lowcarbon-future-we-can-afford-to-go-green.html

[4]Kevin A. Hassett, Aparna Mathur and Gilbert E. Metcalf (2009), “The Incidence of a U.S. Carbon Tax: A Lifetime and Regional Analysis,” The Energy Journal, Vol.30, No.2, March 2009, NBER Working Paper 13554; Aparna Mathur and Kenneth P. Green (2008), “Measuring and Reducing Americans’ Indirect Energy Use, Energy and Environment Outlook, American Enterprise Institute, December 2008; Aparna Mathur and Kenneth P. Green (2009), “Indirect Energy and Your Wallet,” Energy and Environmental Outlook, American Enterprise Institute, March 2009.

[5] Kristian Bodek and John Heywood (2008) “Europe’s Evolving Passenger Fleet: Fuel Use and GHG Scenarios Through 2035.” (MIT: Laboratory for Energy and Environment) Available at http://web.mit.edu/sloan-auto-lab/research/beforeh2/files/Europe’s%20Evolving%20Passenger%20Vehicle%20Fleet.pdf