Looking to buttress their claim that the Environmental Protection Agency's greenhouse gas regulations will do substantial damage to the economy and job growth, Congressional Republicans and conservative media outlets have cited a study of their potential economic effects conducted by Margo Thorning, Chief Economist for the American Council for Capital Formation (ACCF).

For instance, the New York Post's Abby Wisse Schachter recently relied heavily on Thorning's analysis to characterize the EPA rules as “an onerous regulation regime that may result in $75 billion drop in capital formation and as many as 1.4 million jobs lost by 2014.” ACCF's numbers have also been cited by the Senate Republican Policy Committee, the American Petroleum Institute, the Chamber of Commerce and an Investor's Business Daily op-ed in arguments supporting legislative efforts to prohibit EPA from regulating greenhouse gas emissions. However, the study's conclusions are based on an assumption for which it provides no support.

The ACCF analysis, which Thorning presented to the House Energy and Commerce Committee in February, concludes that “if U.S. capital spending declines by $25 to $75 billion, in 2014 there would be an economy wide job loss of 476,000 to 1,400,000 when direct, indirect and induced effects are included. As a result, GDP would be $47 billion to $141 billion less in 2014.” Conservative media outlets that have quoted the study have failed to take note of the word “if” in that sentence. However, the uncertainty underlying the study is important, especially given the fact that Thorning doesn't completely explain the assumptions she uses to get to her estimate of the decline in investment spending (which is in turn used to generate her estimate of job losses.) From Thorning's February testimony [emphasis added]:

When evaluating a prospective investment, business analysts typically add a risk premium to the firm's cost of capital, ranging from 0 to 50% and higher. Assuming that the pending GHG regulations increase the risk premium added to the firm's cost of capital by 30% to 40% and using conservative estimates of the elasticity of investment in response to changes in the cost of capital, it seems likely that U.S. investment could decrease by 5% to 15% over 2011-2014 period compared to the baseline forecast.

[...]

"[A] 5% to 15% decline in investment for only the directly affected industries would result in an approximately $25 to $75 billion reduction in investment outlays."

Thorning's testimony directed readers to further details contained in a declaration she submitted to the DC Circuit Court in September in support of a motion to stay the greenhouse gas regulations. But in that declaration, Thorning provides no additional explanation for the assumption that the EPA regulations “increase the risk premium added to the firm's cost of capital by 30% to 40%.” From the declaration [emphasis added]:

I estimate the current cost of capital for low risk investments is around 6%. However, for investments by entities that will face permitting requirements under EPA's GHG regulations, a substantial risk premium ranging from 30% to 40% would be appropriate. Assuming that the new regulations will increase the cost of capital for firms in many industries besides those classified as energy intensive, from a current average of 6.0 % to as high as approximately 8.5% (or by 40%) and using the more conservative lower elasticity number (-0.25), I estimate that U.S. investment could decrease by between 5% to 15%. The increase in capital costs is likely to result in a reduction in private investment in the U.S. of between $97 and $290 billion dollars in 2011 and $100 to $301 billion dollars in 2014. Losses in private investment of this magnitude would rival those the U.S. experienced since the fourth quarter of 2007.

In fact, other economists question the base assumptions of the ACCF study.

When asked about Thorning's “risk premium” figures, MIT economist and Director of the Brookings Institute's Hamilton Project, Michael Greenstone said, “To the best of my knowledge, that is not a standard assumption.”

Nathaniel Keohane, then-chief economist at the Environmental Defense Fund and now a member of the White House National Economic Council, responded to Thorning's declaration, writing in his own statement to the Circuit Court that although the risk premium assumption “is the crucial step in her chain of calculations,” she does not provide “any evidence to support it”:

Thorning simply asserts that a “risk premium” of 30% to 40% would be “appropriate.” This estimate has no basis in fact; nor does Thorning provide any evidence to support it. Yet this claim is the crucial step in her chain of calculations claiming economic harm. Since her alleged “risk premium” lacks any basis in evidence, the same holds true for her calculated increase cost of capital, and the alleged decrease in U.S. investment -- both of which are based on the supposed risk premium. In summary, Thorning provides absolutely zero evidence for any increase in the cost of capital resulting from EPA's actions, or for any corresponding reduction in investment.

Similarly, Frank Ackerman, another environmental economist, stated via email that Thorning:

assumes that regulation of GHG emissions under the Clean Air Act causes an enormous increase in risk, which increases the rate of profit businesses need to make on investment, which lowers investment, and thereby lowers incomes and employment. She does not cite any independent sources for the huge increase in risk, the engine which pulls this train over the cliff; she just cites other ACCF documents which offer the same off-the-cuff judgment about terrible risks associated with regulation.

Ackerman added: “The rest of her testimony seems to flow more or less straightforwardly from that extreme premise.” Indeed, Thorning takes the figures derived from her “risk premium” assumption and plugs them into a model to get a job impact of 476,200 to 1,428,700. Actually, to be more precise, Thorning doesn't actually appear to have run the model herself. According to the citation provided in her testimony, the “Source” of her job loss numbers turns out to be “Calculations by the American Petroleum Institute using IMPLAN model.”

As for the model itself, Greenstone said: “It's poorly understood but the very notion of counting job losses in the overall economy is something mainstream economists don't believe in.” He said the model is not capable of estimating what happens after the initial reductions in employment due to the regulations and “the notion that affected workers would remain unemployed for the rest of their lives is not credible. However, the evidence suggests that the job losses could lead to wage declines, especially for the workers that formerly worked in the polluting industries,” Thorning's study, however, omits analysis of the regulations' potential effect on wages.

It's also worth noting that the numbers for the decline in investment spending from Thorning's February testimony ($25-$75 billion in 2014) are one quarter of the amount she presented in the September declaration ($100-$301 billion in 2014). She explains in her testimony how she got the lower numbers [emphasis added]:

In the ACCF's initial calculations, submitted to the U.S Court of Appeals for the District of Columbia Circuit, the 5 to 15% reduction was applied to all capital investment in the U.S. economy. These calculations suggested that gross private domestic investment could be reduced by $100 to $300 billion by 2014. Subsequently, ACCF narrowed the focus of the analysis to target the industries included in the EPA's guidelines to determine which specific industries would be impacted with the first wave of GHG regulations under the CAA. Using this methodology, it is estimated that the directly impacted industries, such as the electric power sector, mining, manufacturing and wholesale and retail trade were responsible of 25% of overall capital investment in U.S. economy in both 2008 and 2009. Therefore, a 5% to 15% decline in investment for only the directly affected industries would result in an approximately $25 to $75 billion reduction in investment outlays.

So in September, Thorning applied her “risk premium” assumption to the entire economy, rather than just the regulated industries and said that the EPA regulations could reduce investment by up to $300 billion by 2014 a figure that was widely quoted by EPA's critics. But months later, when she went to produce an estimate for how many jobs would be lost due to the regulations, she put the lower numbers for the decline in investment spending into the model. Why?

Would the higher numbers have produced a job loss figure too high to be believable? Does she think the higher numbers overstate the likely impact or that the lower numbers understate it? I tried to contact Thorning to ask her these questions, but she did not respond to phone or email messages.

Given the lack of clarity surrounding the assumptions behind the study, it's worth considering what, exactly, the American Council of Capital Formation is. According to the group's website, the organization supports “well-thought-out economic, regulatory, and environmental policies to promote capital formation, economic growth, and a higher standard of living for all.” Its board of directors includes former government officials and business figures with major ties to fossil fuels industries, including ConocoPhillips executive and former API head Red Cavaney, Thomas R. Kuhn, president of the Edison Electric Institute, and Dave McCurdy, president of the Alliance of Automobile Manufacturers, along with representatives from at least five lobbying firms -- all of which count among their clients oil, gas, electricity or auto companies.

Additionally, among the few dozen sponsors listed in ACCF's 2010 annual report are:

• American Forest & Paper Association
• American Gas Association
• American Iron and Steel Institute
• American Petroleum Institute
• ConocoPhillips
• Constellation Energy
• Edison Electric Institute
• Exxon Mobil
• Independent Petroleum Association of America
• Koch Companies Public Sector, LLC
• Marathon Oil Corporation
• National Petrochemical & Refiners Association
• National Rural Electric Cooperative Association
• Occidental Petroleum Corp.
• Southern Company
• Sunflower Electric Power Corporation

For years, ACCF has been arming opponents of proposed action to address climate change with studies providing stark and easily quotable figures for the economic harm to come should the govenrment succeed in reducing greenhouse gas emissions.

Accordingly, in 2008, ACCF teamed up with the National Association of Manufacturers to produce an analysis which predicted that Sens. Joe Lieberman and John Warner's climate bill would cost up to 4 million jobs and would lower the average household's disposable income by $6,700 per year by 2030. ACCF's assumptions were cast into doubt at the time as they were far more pessimistic than those used by other studies. The Pew Center on Global Climate Change explained in a May 2008 brief:

Models that constrain the use of potential technologies dramatically increase the costs of reducing emissions. For example, the ACCF/NAM model constrains the future deployment of nuclear energy so that less electricity is delivered from nuclear facilities in the High Cost Scenario than is projected under business as usual forecasts (developed by DOE's Energy Information Administration). Similarly, the amount of electricity delivered by wind power is also constrained to an annual deployment level lower than was actually delivered in 2007. The ACCF/NAM model restricts additional wind capacity to 5 GW/year for the Low Cost Scenario and 3 GW/year in the High Cost Scenario. According to the American Wind Energy Association, there was an additional 5.244 GW of wind capacity added in 2007. The result of these restrictions is that the costs from this model fall far outside the range of other modeling efforts.

In addition, ACCF's study did not fully accountfor provisions of the legislation designed to constrained costs.

Another ACCF/NAM analysis surfaced when the House passed a climate bill in 2009. As The Hill reported, the study, “projected a significantly higher cost to the overall economy from climate legislation than have studies done by the Energy Information Administration (EIA), Environmental Protection Agency (EPA) and Congressional Budget Office (CBO).” The basis of ACCF's projected economic impacts were based on familiar pessimistic assumptions. As Gristreported:

The NAM study predicts that just 10 to 25 gigawatts of new nuclear power would be developed under the bill. But the EIA estimates that 11 additional gigawatts of nuclear power would come online by 2030 without a cap on carbon, and up to 135 gigawatts under the Waxman-Markey bill.

There are also some dubious assumptions about offsets. While the Waxman-Markey bill allows for up to 2 billion tons of offsets -- half domestic, half international -- the NAM study assumes that 95 percent of those offsets would be domestic. Domestic offsets are far more expensive than international offsets.

The National Journal reported on October 31, 2009, that “business lobbyists have relied greatly” on the ACCF/NAM study.

ACCF released yet another study for the Kerry-Lieberman climate proposal in July 2010, by which time the legislative effort had essentially disintegrated.