NERA's flawed critique of EPA's analysis of the Clean Air Act

The Environmental Protection Agency’s (EPA) cost benefit analysis of the Clean Air Act is probably one of the most peer reviewed and thorough cost benefit analysis ever carried out by an institution, be it government, private, or academic. But opponents of clean air and public health are intent on discrediting it (click here to read about the Heritage Foundation’s critique a few months ago, and documentation of EPA’s peer review process), because the gains to public health from reducing pollution swamp the costs. For 2010, EPA estimated that benefits exceeded costs by a ratio of almost 26 to 1, and by 2020 30 to 1. In a review of federal regulations, the Office of Management and Budget, a tough critic of EPA, similarly concluded that benefits from clean air regulations were much higher than costs.

A new report by National Economic Research Associates (NERA), prepared by Senior Vice Presidents Charles Montgomery and Anne Smith, again attempts to invalidate EPA’s analysis. NERA and the two authors have been in the business of anti-regulatory industry studies for a long time. Over the years, the company has conducted numerous analyses at the behest of polluters (e.g. click here, here and here for studies conducted for the National Association of Manufacturers, the American Petroleum Institute, and the Electric Power Research Institute, respectively). More recently, NERA produced a state-by-state analysis against proposed ozone standards (click here and here for critiques of a study that relied upon this analysis, by Professor Richard Howarth of Dartmouth College and myself, respectively). As former Vice Presidents of Charles River Associates, Montgomery and Smith used to conduct similar analyses against regulations (click here for CRA’s analysis of the Climate Security Act of 2007 supported by the American Petroleum Institute and Edison Electric Institute, and here and here for their analysis (funded by the National Black Chamber of Commerce) of the American Clean Energy and Security Act of 2009 and my critique of it, respectively).

The analysis

Montgomery and Smith argue that EPA’s estimated benefits of reducing risks to life from regulating pollution, which account for the lion’s share of benefits, are meaningless because they do not represent an actual good exchanged in markets and therefore are unrelated to real output and gross domestic product (GDP). (GDP is the market value of all final goods and services produced in the economy, as measured by their prices. It represents payments to all factors/costs of production, such as labor and raw materials). The authors further assert that because the monetized value of the benefits equate to approximately 10% of GDP, the cost benefit analysis cannot possibly be right, as surely people would not give up 10% of GDP to have them.

These arguments are premised on flawed economics and defy common sense. The authors make three errors:

• First, they confuse GDP with well being. As is widely recognized in the economics profession, GDP includes some activities that don’t reflect net increases in welfare, such as the goods and services needed to rebuild infrastructure after a natural disaster. It also does not count many activities that do provide net increases in welfare, such as the intrinsic value people place on nature (e.g. national parks, coral reefs, or rare species). Do these things not exist? Is cleaner air a figment of our imagination?
• Second, the authors ignore the fact that GDP does not include damages to people who are not compensated for having to breathe polluted air. Economists call these damages “negative externalities,” to reflect the notion that they are not “internalized” in the market price of the good whose production is causing pollution. A proper cost benefit analysis therefore estimates what the value of these externalities would have been had there been no regulation, and counts the cleaner air from regulation as a real benefit (in the absence of regulation, pollution would be counted as a production cost in GDP if it were "internalized" in the market price). Because economists do not have actual prices to work with, they must estimate them using as proxies either what people would be willing to accept to incur the harm, or what they would be willing to pay to avoid it. For this analysis, EPA uses the former measure, approximating the value of reduced risks to life from pollution reduction using studies that estimate how much more workers require in compensation for given increases in mortality risks on the job*** (controlling for other factors related to wage levels, such as skill level, education, etc).
• Third, and following from the previous point, it is incorrect to say that EPA’s estimated benefits are not real and are unrelated to GDP simply because there is no market price for them, or because people would not be willing to sacrifice 10% of GDP for cleaner air. Aside from the obvious point that we did not have to give up 10% of GDP for the cleaner air we now have (we have both the measured level of GDP and the benefits of the Clean Air Act), how (and whether) property rights are defined can make the difference between whether a good (or “bad” in the case here) ends up being reflected in markets, and therefore counted in GDP. If one assumes that the person breathing the pollution has the right to breathe clean air rather than the polluter the right to make it dirty, and that markets are capable of enforcing this assignment of property rights (for various reasons, they rarely are, hence the need for regulation), the polluter would have to pay the victim for the right to pollute, and this would enter the GDP calculation as a cost of production. Wealth and goods would be redistributed away from polluters toward individuals suffering from the pollution who now “own” it as if it were any other resource. How much would the monetized value of the pollution costs equate to in the GDP calculation? I myself would speculate that, under such a property rights assignment, the amount people would require to “sell” clean air to polluters might approach oh, say, 10% of GDP?...Whatever the amount, it would probably exceed the amount sellers could recoup in product prices. Goods generating excessive and dangerous pollution levels would no longer be profitable to make. Furthermore, producers would have an incentive to invent new, less polluting, goods, and to make existing products more cleanly. It follows from these propositions that the supposedly meaningless benefits EPA estimates would not have to be calculated in the first place…

Conclusion

To say that EPA’s analysis is invalid because some of the health benefits it estimates have no market price and are therefore not related to GDP, or because people wouldn't be willing to give up 10% of GDP for cleaner air, is wrong and misleading. It assumes that GDP accurately measures well being, and overlooks the obvious: we already have both the measured level of GDP and cleaner air. The authors present a false choice. They also implicitly assume that GDP does not include pollution costs if property rights are clear and enforceable, in this case if polluters are required to compensate those who have to breathe the dirtier air. Under these circumstances, the pollution would be included in GDP as a cost of production. Finally, there would be strong incentives to make products more cleanly, and invent ones that minimize pollution levels. Pollution would be lower, and EPA would not have to estimate what Montgomery and Smith claim are “meaningless” benefits.

Overall, the authors’ critique of EPA’s cost benefit analysis is not supported by economic theory or common sense. The fact that basic tenets of economics that undermine their argument are missing should not be surprising: the industries that regularly hire them are legally obligated to maximize profits for shareholders. As we know all too well, this can come at the expense of the health and well being of everyone else.

-------------------------------------------------------------------------------------------------------------- 

*** A few studies EPA reviewed in estimating the value of reducing risks to life were based upon hypothetical surveys, but the final value used was approximately equal to the central value of the wage-based studies, of which there were many more (Montgomery and Smith also try to dismiss the benefits by saying they were based on hypothetical data). I should also note that there are a number of factors that could make the amount people are willing to accept to incur pollution-induced risks higher than what they would accept for job-induced risks. If the reader is interested in these factors, feel free to contact me to discuss them.

Updated August 12, 2011 to include link to NERA report.

Comments