Document Type

Working Paper

Publication Date



Richard Paul Richman Center for Business, Law, and Public Policy


The Charles Evans Gerber Transactional Studies Center


The government often relies on targeted subsidies for both “green” energy and hydrocarbons. These subsidies pursue worthwhile goals. But unfortunately, many have design flaws that make them less effective or even counterproductive. The goal of this Article is to show how we can do better.

This Article focuses on three sets of issues. First, there often is tension between our environmental and national security goals. Unfortunately, the economics literature on energy largely ignores these tradeoffs by omitting national security from the analysis. This Article takes issue with this approach and suggests ways to manage these tradeoffs. Second, as a strategy to deal with these and other costs, this Article criticizes subsidies, and argues that Pigouvian taxes are a better alternative. Third, if we are stuck with subsidies for political reasons, this Article offers a number of ways to improve them.

The U.S. faces increased defense costs in securing access to oil, for instance, in deploying forces in the Middle East. Yet the economics literature on energy does not account for these national security costs, arguing that they are too difficult to quantify. Even so, difficulties in computing costs are not a reason to ignore them. This Article argues that national security costs need to be incorporated in the analysis, and considers ways to manage tradeoffs between these costs and environmental externalities.

The best way to manage these tradeoffs is with Pigouvian taxes. For instance, we should have a national security tax, a carbon tax, as well as a pollution tax. But instead of this menu of Pigouvian taxes, the U.S. has relied on targeted subsidies for particular technologies. While other commentators have criticized this choice, this Article offers a different critique, based on administrative costs, that is new to the literature. Since producers and consumers can increase (or decrease) negative externalities in a multiplicity of ways, the cost (or reward) should be the same, regardless of how they do it. But this Article shows why consistency is easier for taxes than for subsidies. While taxes have to reach all sources of harm, subsidies have to reach all ways to mitigate harm, which is a much longer list. To be comprehensive, we would have to subsidize driving more efficient vehicles, driving less, driving slowly, accelerating evenly, pumping up the tires, carpooling, taking mass transit, and much more. As a practical matter, this is not feasible.

If we are stuck with subsidies for political reasons, this Article offers a number of ways to improve them. For example, subsidies should be precise in rewarding the desired behavior, instead of a proxy for it. Otherwise, the subsidy can have perverse effects. For instance, Congress pays 2.3 cents for each kWh of electricity generated with wind. To claim this subsidy, producers sometimes generate electricity that no one needs, and then pay customers to take it. This perverse effect arises because this subsidy rewards a proxy (producing wind energy), instead of the desired result (replacing carbon-based energy).