Environmental Economics

Carbon Abatement Policies and Technological Innovation

  • Abstract
    This project will analyze a variety of policy issues raised by the dynamic interaction between technological innovation and reductions in carbon emissions. In particular, we will assess the cumulative costs of carbon abatement policies, taking into account their impact on technological innovation, and the costs of R&D activity. Other questions to be addressed include whether is it optimal to conduct a lot of R&D into cleaner technologies now, or to smooth R&D activity out over time. Whether emissions reductions should be delayed until future technological innovation has significantly reduced the instantaneous cost of emissions abatement. Why emissions taxes and tradeable emissions permits differ in their effectiveness at stimulating technological innovation, and how empirically important these differences are. How inefficiencies in the patent system affect the incentives for R&D into cleaner technologies and how they affect the optimal quantity of carbon permits, or optimal level of carbon tax, over time. The analysis will be developed in three stages. First, the optimal trajectories for the level of instantaneous carbon emissions abatement, and the optimal investment in “environmental” R&D, will be characterized over future periods using a social planning model. Second, we will examine the effects of tradeable carbon permits and carbon taxes in a decentralized version of the model and their interplay with induced innovation. At this stage we will ignore imperfections in the patent system. The costs of achieving particular objectives for future emissions reductions under carbon taxes and tradeable carbon permits will be compared. Third, the analysis will be extended to incorporate a model of competition for patents for carbon-reducing technologies. This introduces various imperfections, such as the suboptimal diffusion of new technologies due to the monopoly power created by patents, and problems caused by imitation around patented technologies. At each stage of the analysis we will begin with analytical models to establish qualitative results. In addition, quantitative results will be obtained by calibrating the models to existing estimates of the cost of emissions abatement, and assuming different scenarios for the effect of R&D on reducing economy-wide carbon emissions. To obtain analytical solutions to the model will require using quadratic functional forms. These models will be extended to allow for more general functional forms, and solved either by taking analytical approximations or by numerical simulation techniques. Throughout the project, we aim to provide highly intuitive qualitative and empirical analysis from relatively straightforward models, rather than detailed computational analysis. The proposed research will improve our understanding of how technological innovation may affect the cumulative cost of future policies to reduce U.S. carbon emissions. Many of the insights will be of relevance to the regulation of other pollutants.
  • Metadata
    Principal Investigators:
    Parry, Ian
    Technical Liaison:
    Research Organization:
    Resources for the Future
    Funding Agency/Program:
    Grant Year:
    Project Period:
    September 30, 1997 to September 29, 2000
    Cost to Funding Agency:
  • Project Reports

    Objective: The objective of this research was to analyze the impacts of alternative environmental policies on the incentive to develop cleaner production technologies over time, and what this implies for the appropriate choice among policy instruments and the appropriate stringency of regulations. The focus was mainly on market-based instruments of the type that might be employed to reduce carbon emissions in the United States: grandfathered emissions permits, auctioned emissions permits, and emissions taxes.

    The research addressed two key issues. First, the research investigates the magnitude of the welfare gains from innovation induced by environmental policies. Knowing the magnitude of these welfare gains would be particularly useful as it would reveal to what extent previous cost/benefit studies based on exogenous technology models might have understated the overall welfare gains from environmental policies. Moreover, whether the welfare gains from induced innovation are large or not affects how important a consideration innovation incentives should be in the design of environmental policies. Second, the research compares the efficiency effects in the R&D market induced by different environmental policies, taking into account various imperfections in the R&D market such as the inability of innovators to fully capture the spillover benefits of new technologies to other firms. In this way, we illustrate why, and under what conditions, the welfare discrepancies between the emissions control policies are significant, and other cases when the welfare differences are relatively unimportant from a practical policy perspective.

    The project produced three main theoretical studies of the economic impacts of environmental policies on technological innovation, two of which are published or forthcoming in scholarly journals, and the other of which is under review. In addition, a review essay of the issues will be published in 2002 in a scholarly journal, and a couple of shorter articles also have been published.

    Summary/Accomplishments: When evaluating the costs and benefits of environmental policies, economists have traditionally used static models in which the state of technology is fixed. However the state of technology is endogenous over the long run, and in the absence of policy intervention the amount of effort into developing cleaner production technologies for the future is likely to be suboptimal due to a "double externality" problem. First, if firms do not have to worry about pollution control they lack incentives to develop cleaner technologies. Second, even if they did invent cleaner technologies they may be unable to capture spillover benefits to other firms, due to the public good nature of knowledge. This means that in a dynamic context, environmental regulations can improve economic efficiency not only by limiting pollution, but also by stimulating the innovation of cleaner technologies, which would otherwise be suboptimal.

    There is theoretical literature on the effects of different environmental policies on the incentive for firms to innovate and adopt cleaner production techniques. However, what matters for the net benefits to society from environmental regulations is not the impact on innovation itself, but rather the welfare gains from induced innovation, that is, the benefits of innovation less the costs of research effort. One contribution of this project was to develop a series of models that account for research costs and therefore can pay explicit attention to the welfare effects of environmental policies. Moreover, most of the prior literature has been qualitative in nature rather than quantitative. Another central feature of this project was to conduct extensive simulation exercises to illustrate the magnitude of the welfare gains from induced innovation and how large they might be relative to welfare gains from pollution control that economists have paid attention to previously.

    Knowing the size of the welfare gains from innovation is useful for a number of reasons. First, it indicates to what extent previous, exogenous-technology models may have understated the net benefits to society from environmental regulations. Second, it may shed some light on the quantitative importance of previous theoretical analyses demonstrating that different emissions control instruments provide different incentives for technological innovation?the larger the welfare gains from innovation the stronger the case for preferring instruments that are more effective in promoting innovation. Third, if the potential welfare gains from induced innovation are large, this may provide an efficiency argument for buttressing environmental policies with other instruments to address the problem of technology spillovers such as research subsidies, tax credits, and technology prizes.

    In the paper "How Important Is Technological Innovation in Protecting the Environment?" we assess how large the economic gains from stimulating the development of cleaner production technologies might be. We provide a general treatment of this issue, using a dynamic optimization model in which the control variables in each period are the amount of pollution abatement and the amount of R&D investment. R&D enhances a knowledge stock, which reduces the costs of future abatement. The analysis allows for flow pollutants (e.g., sulfur dioxide, nitrogen oxide) and stock pollutants (e.g., carbon dioxide), and for linear and convex environmental damage functions. The paper makes use of both analytically solved models to provide intuition, and numerically solved models to gauge the quantitative importance of technological innovation.

    The analytical model shows that the relative welfare gains from innovation boils down to three key factors?the initially optimal level of abatement, the speed at which innovation reduces future abatement costs, and the discount rate. We conduct extensive simulation exercises to illustrate a broad range of different scenarios for these factors.

    In general, we find that the economic gains from innovation are significant, but in most scenarios they are less important than the economic gains from curbing excessive amounts of pollution over time. This result was initially a little surprising to us, but it makes sense when we think about the factors that limit the potential economic gains from innovation. First, there is simply a limit to what innovation can achieve?at best it can reduce pollution abatement costs by 100 percent. We show that, even if innovation reduces abatement costs by a substantial amount over time, the gains from this are not necessarily that much larger than the gains from achieving the initially optimal amount of pollution control. Second, the benefits from R&D occur in the future. It takes a while, maybe 10 years, for firms to invent and then adopt major new techniques for pollution control. In contrast, the benefits from reducing pollution occur today. So when we discount future benefits, this reduces the gains from current R&D activity relative to the gains from current pollution control. Third, innovation is costly rather than free. Firms have to pay for scientists, engineers, research laboratories, and so on to develop new technologies. These costs have to be subtracted from the benefits in order to obtain the net economic gains from innovation

    The one case when the economic gains from innovation could be relatively more important than the economic gains from pollution control is when abatement costs are initially "large" relative to environmental damages, and there is potential to realize fairly rapid cost savings from innovation (about a 50 percent reduction in abatement costs in 10 years). Our results apply equally for flow and stock pollutants, and for linear and convex environmental damage functions. In general, they also are robust in comparing economic gains from innovation and abatement over shorter periods within the planning horizon.

    The paper "Instrument Choice for Environmental Protection When Technological Innovation is Endogenous" presents an analytical and numerical comparison of the economic gains from alternative instruments for environmental protection in the presence of endogenous technological innovation. We focus on emissions taxes and both free (grandfathered) and auctioned emissions permits. The paper shows that under different sets of circumstances, each of the three policies may induce a significantly higher net gain than the other two policies. In particular, the relative ranking of policy instruments can crucially depend on the ability to appropriate innovation rents, the costs of innovation, the slope and level of the marginal environmental benefit function, and the number of firms producing emissions. Moreover, although in theory the economic gains from policies differ in the presence of innovation, sometimes these differences are empirically small. This is more likely when firms anticipate that policies will be adjusted over time in response to innovation. In short, there is no clear-cut case for preferring one policy instrument over another.

    The paper builds on earlier literature in a number of respects. First, the paper provides a quantitative comparison of policy instruments. This is important for identifying the circumstances when there can be significant discrepancies in net gains between the policy instruments and where?for practical purposes?these discrepancies are unimportant. Most previous studies have been theoretical and therefore do not explore whether the differences between policies are significant or not. Second, the analysis focuses on net economic benefits, that is, it ranks policies on innovation benefits net of the costs of innovation. In contrast, most previous studies rank policies according to how much innovation they might induce. However, it is important to look at the net gains; for example, when the marginal environmental damage function is very steep we find the net gain to society can be lowest under emissions taxes, even though they induce the highest amount of innovation. Third, the paper provides a more rigorous treatment of the technology adoption decision at non-innovating firms. In particular, we allow for non-innovating firms to have the option of choosing alternative technologies to the new innovation (such as imitations of the new innovation). This modeling extension shows that, at least for competitive industries, previous theoretical work may have greatly overstated the innovation incentives from auctioned emissions permits.

    The first paper of the project, "Pollution Regulation and the Efficiency Gains From Technological Innovation," uses analytical models and simulations to demonstrate two main points. First, prior studies suggest a preference for emissions taxes over (non-auctioned) emissions permits and performance standards based on their potential for promoting technological innovation. The paper casts some doubt on the empirical importance of this assertion: it shows that the economic gain induced by an emissions tax is significantly larger than that induced by other policies only in the case of very major innovations. Second, the paper also finds that the presence of technology spillovers per se does not necessarily imply large inefficiencies. In part, this is because competition for a given amount of innovation rent can be excessive and this tends to offset, to some extent, the inadequate incentives due to imperfect appropriation of innovation rents. Thus, despite spillovers, the net economic gain from additional policies to promote innovation (such as R&D subsidies) may be limited.

    Publications and Presentations: Total Count: 11

    TypeCitationJournal Searches
    Journal ArticleFischer C, Parry I, Pizer W. How important is technological innovation in protecting the environment? American Economic Review 2000. not available
    Journal ArticleFischer C, Parry I, Pizer W. Instrument choice for environmental protection when technological innovation is endogenous. Journal of Public Economics. not available
    Journal ArticleFischer C, Parry I, Pizer W. Instrument choice for environmental protection when technological innovation is endogenous. Journal of Environmental Economics and Management. not available
    Journal ArticleParry I. On the implications of technological innovation for environmental policy. Environment and Development Economics. not available
    Journal ArticleParry I. Pollution regulation and the efficiency gains from technological innovation. Regulatory Economics 1998;14:229-254. not available
    PaperFischer C. Climate change policy choices and technological innovation. Issue Brief #20, Resources for the Future, Washington, DC, 2000. not available
    PaperFischer C, Parry I, Pizer W. How important is technological innovation in protecting the environment? Discussion paper 00-15, Resources for the Future, Washington, DC, 2000. not available
    PresentationParry I, Pizer W, Fischer C. How important is technological innovation in protecting the environment? Presented at Resources for the Future, May 2000. not available
    PresentationFischer C, Parry I, Pizer W. Instrument choice for environmental protection when technological innovation is endogenous. Presented at a Conference on Major Technological and Environmental Risk, Montreal, Canada, November 1998. not available
    PresentationParry I, Fischer C. Instrument choice for environmental protection when technological innovation is endogenous. Presented at the National Bureau of Economic Research Workshop Cambridge, MA, August 1998. not available
    PresentationParry I. Pollution regulation and the efficiency gains from technological innovation. Presented at Stanford University, April 1998. not available

    Supplemental Keywords: clean technologies, innovative technology, environmentally conscious manufacturing, cleanup, public policy, cost benefit, modeling, analytical, global climate, sulfates, nitrogen oxides.
  • Project Status Reports