What’s the Best Way to Promote Green Power?

Right now, the only thing that competes in the news cycle besides the war in Iraq and the upcoming election is green power, in all its forms. Consumers are buying compact fluorescent light bulbs, utilities are promoting their latest options, and the "experts" are claiming that their favorite source should prevail. And renewable energy sources—including geothermal, solar, wind, tide and hydropower—are a major component of most strategies for addressing global climate change. As it turns out, not all policies that promote renewables are created equal; our research shows that broad-based policies like emission fees are substantially more cost effective than more targeted approaches, such as research and development subsidies, if the goal is to reduce greenhouse emissions in the near and medium terms.

Economists typically argue that a direct price for CO2 (a tax or tradable emissions permit system) would provide the most efficient incentives for development and use of cleaner technologies. In practice, a number of issues are at play and therefore numerous policies are being pursued. Some of these issues extend beyond economic efficiency, such as political acceptability and distributional questions. Others emerge directly from economic efficiency considerations; for example, spillovers in R&D markets reduce incentives for firms to innovate because a portion of the returns on their investments will be captured by others. Similarly, the innovation process may occur not only through R&D investments, but also through firms "learning" from the production and use of new technologies.

Most prior studies have focused on the effectiveness of emissions pricing policies, such as emissions taxes and emissions permits, for stimulating innovation in green technologies. The broader, more pragmatic set of policies, such as those using performance standards and supporting renewable energy, have generally been neglected.

To shed some light, we developed a modeling framework for the electricity sector that incorporates both a knowledge accumulation stage, when R&D and learning occur, and a knowledge application stage, when the cost-reducing benefits are realized. Using this consistent framework, we evaluated the six aforementioned policy options for their relative performance according to different metrics: emissions reduction, renewable energy production, R&D, and economic surplus. To better understand both the magnitude of the efficiency and cost differences among the policies, we applied our approach to a numerical model of the U.S. electricity sector.

We set an emissions price of $7 per ton of CO2 (or about $25 per ton of carbon) throughout the model and used the resulting emissions as a target for the other scenarios, allowing for an apples-to-apples comparison. For the portfolio standard and the emissions performance standard, we held the price of credits constant across our two time stages, while meeting the implied emissions target. The resulting renewables portfolio standard rises from 6.0 percent in the first stage of the model to 9.6 percent in the second stage, which is close to a recent proposal for a national renewables portfolio standard that would rise from 5 percent by 2012 to 10 percent by 2020.

Our results indicate that the emissions price is indeed the most efficient means of achieving a given emissions target, leading to the least cost in terms of surplus and requiring the least investment in renewable energy R&D. Conversely, the renewables research subsidy is by far the most costly single policy for reducing emissions.

The figure on page 12 displays the relative costs of the policies as a ratio to the cost of the emissions price. An optimal combination of policies performs best and actually leads to a small cost savings with this modest emissions target due to induced innovation benefits. The renewable portfolio standard is roughly twice as costly as the emissions price, with the performance standard and the output tax lying in between. The renewables production subsidy is two-and-a-half times as costly, while relying on the R&D subsidy alone is a whopping 12 times more costly than utilizing a price on emissions.

When the ultimate goal is to reduce emissions, policies that also create incentives for fossil-fueled energy generators to reduce emissions intensity and for consumers to conserve energy perform better than those that rely solely on incentives for renewable energy producers. For the modest emissions targets we examined, a renewable energy R&D subsidy turns out to be a particularly inefficient means of emissions reduction—when used alone—because it postpones most of the effort to displace fossil-fueled generation until after the costs of renewables are reduced.

Although climate change is a long-term problem, our results for mid-term strategies emphasize the important role of policies that encourage abatement across all forms of energy generation and time frames, as well as the limitations of narrowly targeted policies.

Nonetheless, no single policy can simultaneously correct more than one market failure—in this case the failures associated with the emissions externality and the knowledge spillovers from learning and R&D. Each policy poses different trade-offs. In the presence of knowledge spillovers, an optimal portfolio of policies—an emissions price combined with optimal learning and R&D subsidies—can achieve emissions reductions at significantly lower cost than any single policy alone. (Yet the emissions reductions continue to be attributable primarily to the emissions price and the learning subsidy is small.)

If even a modest emissions price is not politically feasible, an R&D subsidy by itself is not the next best policy, and the costs of that political constraint are likely to be quite large and increasing with restrictions on the remaining policy options. It should be kept in mind, however, that we chose to focus on reductions over the near- to mid-term and incremental improvement of existing technology, rather than the development of breakthrough technologies that might achieve deep reductions. R&D policies probably have greater salience in the latter context, but that should not diminish the role of emissions pricing to improve the competitiveness of all green alternatives in the market.

This article is based on a longer work by the authors, "Environmental and technology policies for climate mitigation," Journal of Environmental Economics and Management, 2008, 55(2): 142-162.