In the previous chapters we studied pollution problems mainly in a static framework. With noise pollution, for example, the externality results directly from the instantaneous flow of noise. Many other pollutants, however, accumulate, and the externality results from the stock of the pollutant rather than from its flow. This is in particular the case for greenhouse gasses such as carbon dioxide. It is also relevant for air pollutants such as sulfur dioxide and nitrogen oxides, as well as for liquid pollutants that accumulate in groundwater reservoirs. This is not to say that our analysis so far is valid only for noise or other non-accumulating externalities, in that we can interpret our earlier findings as steady-state results. In this chapter, however, we focus directly on the dynamic consequences of accumulating pollutants. Although we do not deal in depth with resource economics in this book, pollution abatement for accumulating pollutants has parallels to the exploitation and management of renewable resources. In the case of climate change, for example, the atmosphere can be thought of as a sink for pollutants, and as such as a renewable resource.

To analyze accumulation dynamics, we need additional analytical tools from dynamic systems analysis. When conducting analysis in continuous time, we typically apply methods from dynamic optimization and optimal control. A full introduction to optimal control theory is beyond the scope of this book. Nonetheless, in appendix A to this chapter we briefly define the main concepts of optimal control theory, and lay out in a recipe-like fashion the corresponding optimality conditions.

This chapter is organized as follows. In section 13.1 we set up a dynamic version of our Chapter 3 model, with environmental damage resulting from accumulated pollution. We start by briefly presenting a discrete time version of the model to derive the optimality conditions via the familiar Lagrange method. In section 13.2 we then analyze the continuous time model in detail by first setting up the optimality conditions and looking at the optimal steady state, and then turning attention to the optimal path into the steady state. We do this analytically and using phase diagram analysis. In section 13.3 we link the problem of an exhaustible resource to the problem of an accumulating pollutant.

We saw in Chapter 21 that cost-benefit analysis (CBA) is a technique used to assess whether a project is “worth it” from an efficiency perspective. As the name implies, a critical step in cost-benefit analysis is the estimation of the economic costs and benefits of the proposed action. In this chapter, we continue our discussion of CBA for environmental policy by focusing on two estimation-related topics: benefits transfer, and the empirical characterization of abatement cost functions.

We have noted that the benefits of environmental policy outcomes are usually non-market, meaning the methods presented in Part III of this book are critical inputs for executing at least part of an environmental cost-benefit analysis. In the context of an original study, however, non-market valuation methods require a non-trivial commitment of resources and often the skills of a specialist. This observation often presents policy analysts with a two-fold dilemma. First, in some instances deadlines and institutional constraints make execution of an original valuation study in support of a specific policy analysis infeasible. Second, many decisions – particularly those taken at more localized levels – are too limited in scope to justify expenditure on original research. In these cases, practitioners of CBA need some method of assessing non-market benefits without resorting to an original study. That is, the benefits of the proposed policy need to be measured using previous studies and secondary sources of information. The art and science of doing this is known in environmental economics as “benefits transfer.”

In contrast to benefits, it is often thought that the costs of environmental policy are relatively straightforward to estimate. For example, observation of accounting records on what regulated entities spend to comply with a policy should provide the data needed to tally the cost side of cost-benefit analysis. While this may be a useful first step, our discussion in Chapter 5 demonstrated that abatement cost, when properly viewed as the opportunity cost of reducing emissions, is a broader concept than firms’ compliance costs. Thus far in the book, however, we have said little about how to translate the conceptual aspects of abatement cost functions into empirical strategies for measuring the costs of environmental policies.

Throughout Parts I and II of this book we have referred to damage, abatement cost, profit, and other economic functions in our discussion of environmental policy. As we have seen, these functions arise from household preferences and firm production technologies, and are therefore usually not directly observable. Nonetheless, to this point we have assumed regulators and other agents possess knowledge of these functions or estimates thereof, and conditional on this, we have proceeded with our mostly theoretical discussion of policy design. In Part III, we turn our attention to methods that have been developed to estimate the damage and cost functions needed to implement policy. With this, our emphasis switches to examining models of individual and household behavior, since the measurement concepts of interest usually arise out of individuals’ interactions with the environment. Furthermore, Part III of the book has a greater emphasis on econometrics, since our discussion will also examine ways that these models can be brought to data to produce empirical results.

The themes we discuss in Part III are usually referred to collectively as the sub-field of non-market valuation. The general problem of non-market valuation is to compute a monetary value for environmental resources or services that, in the general setup of market economies, are not subject to exchange, and therefore do not have an observable market price. For example, we might be interested in knowing the monetary value of preserving a wooded area in an urban landscape, relative to its value in a residential use. Likewise, large-scale biodiversity preserves usually preclude extractive use, such as mining and timber harvest, within the boundaries of the preserve. What is the monetary value of the biodiversity protection relative to the market value of the forgone minerals and lumber? To answer this and similar questions, conceptual models that connect the environmental resource to people's preferences (or in some cases firm profits), and econometric approaches that connect data to the conceptual models, are needed.

For this our starting point is neoclassical welfare theory, as it was when we started the book. This helps us to stress that the valuation problem should not be viewed as separate from the policy design problem in Part II, since they are two sides of the same coin.

To this point in Part II of the book we have generally considered the regulation of point sources, and with the exception of the spatial problem in Chapter 8, have assumed there is a one-to-one relationship between emissions and the ambient pollution level. This allowed us to measure environmental damages as a function of total emissions. In addition, we have implicitly assumed that emissions from particular sources are perfectly observable, so that responsibility for a given emission level can be assigned to a specific source. In this chapter we turn our attention to ambient pollution problems, in which environmental damage is assessed based on the concentration of effluent in the environment rather than the count of emissions at the polluting source. The ambient pollution problem has two crucial characteristics. First, ambient pollution problems often arise from non-point sources of emissions. A prominent example is agriculture, where nutrient fertilizers, animal waste, sediments, and pesticides can run off fields or leak from waste management systems into surface and groundwater. Emissions of these substances from individual agricultural producers are usually not observable. Rather, the collective consequences of all farmers’ emissions in a watershed are reflected in ambient concentrations of effluent in lakes, streams, and wells. Second, in many cases there is not a deterministic relationship between total emissions and ambient pollution problems. Instead, the relationship is stochastic, with randomness coming from weather events and other exogenous shocks beyond the control of polluters and the regulator. Once again considering the agricultural example, nutrient concentrations are generally higher in streams when the water volume is low. Similarly, heavy rains can increase the amount of nutrient and sediment runoff, thereby causing acute spikes in ambient pollution levels.

In this chapter we discuss the challenges associated with regulating ambient pollution levels. We begin in the next section by focusing on the non-point source pollution problem in a deterministic environment, which illustrates how moral hazard – i.e. the inability of the regulator to observe individual polluters’ emissions – necessitates the use of an ambient pollution tax in lieu of an emission tax. In section 9.2 we introduce a stochastic relationship between emissions and ambient pollution. We show that, under constant marginal damage, a uniform ambient tax can induce the efficient abatement effort, while under increasing marginal damage a uniform tax is only second-best optimal.

Human history has been defined by innovation and technological progress. Economists have for decades recognized that technological progress is critical for increasing general material wealth and the quality of human life. More recently, economists began focusing on the specific role that technological progress plays in the environmental context. In this chapter we synthesize the large conceptual literature related to innovation, technology adoption, and the environment.

The general topic of technology nests several related but distinct sub-areas of environmental economics, and so in this chapter we will look at multiple decision margins from a variety of actors. For example, when subject to environmental regulation, polluting firms can directly invest in R&D (research and development) efforts to reduce their abatement costs. In section 11.1 we study this type of behavior when investment has purely private benefits, and ask if additional efforts by the regulator are necessary to assure an efficient amount of investment in abatement cost reduction. In section 11.2 we generalize the model to allow for investment spillovers, whereby R&D effort by an individual firm creates knowledge that serves to reduce the abatement costs of all firms in the industry. We will show that, while purely private benefits from investment do not result in additional market imperfections, the spillover case produces a second market failure, based on the public good nature of R&D effort. Thus two policy instruments are needed to reach an efficient outcome.

In other contexts, polluting firms do not internally innovate, but instead must decide whether to bear the fixed cost of installing an available (abatement) cost-reducing new technology. We study this context in section 11.3, where we compare private and socially optimal adoption outcomes and discuss the performance of our suite of policy instruments in regulating emissions, and encouraging firms to deploy the new technology. The incentives for technology adoption that the various policy instruments provide have been loosely referred to in the literature as their “dynamic efficiency” properties. We discuss rankings of policies relative to a common environmental starting point, as well as their comparative usefulness for decentralizing the fully efficient outcome.

Our discussion of environmental policy to this point has generally abstracted from the many ways that institutional and design elements can affect the performance of the suite of policy instruments. For example, while we have distinguished between auctioned and freely distributed permits in cap and trade programs, we have not addressed the many ways that freely distributed permits can be assigned to participating firms. In this chapter we discuss a collection of institutional issues in cap and trade programs, analyzing how design elements that are common in practice may affect performance.

We focus specifically on cap and trade because its theoretical merits have long been stressed by economists, and these efforts have paid off in the sense that emission trading is now part of the common parlance in environmental policy discussions. Indeed we now have experience with two major programs – the US sulfur dioxide program, and the European Union Emissions Trading System for carbon dioxide – and several smaller, regional programs. It is telling, however, that few if any of the existing programs were designed according to the principles described in the theoretical papers. Instead, they combine political compromises, interest group demands, and economic ideas in ways that have on occasion contributed to disappointing performances for cap and trade programs.

In what follows we examine seven topics related to cap and trade programs, and which economists have examined. In section 8.1 we present an analysis that explicitly accounts for differences in the location of polluting firms and the spatial extent of the damages their emissions cause. We show that cap and trade programs can be designed to accommodate the spatial dimension, but that it involves a change in focus from emissions to ambient pollution outcomes. In section 8.2 we address the temporal dimension by considering the merits of allowing polluters to transfer emissions across time by permitting them to bank (or borrow) emissions rights. Both the US SO2 program and the European Union CO2 program allow some form of permit banking, and so understanding how this affects behavior in these programs is necessary for evaluating their success. We show that, under some circumstances related to abatement cost uncertainty, some limited forms of banking are welfare improving.

In the last chapter we introduced the output market to our analysis and used it to extend several of the results we derived in Chapter 3. Throughout we maintained the assumption that all markets were competitive, which allowed us to show that the optimal level of pollution was still characterized by the marginal abatement cost equals marginal damage condition. In this chapter we consider how optimal policy might be different under non-competitive market structures. As we will see, our analysis is complicated by the presence of both market power and pollution-related distortions, and these will cause some of our familiar results to no longer hold. Indeed, the presence of multiple distortions means the regulator's objective is typically to locate a second-best optimal emission tax rate that reflects the relative strength of the two market distortions. The material we present in this chapter will therefore be more technical than what was used in Chapter 5.

What is our motivation for including a chapter on non-competitive markets? Although competitive markets provide a useful starting point for studying environmental policy, non-competitive industries are the primary sources of many regulated pollutants. For example, in both the United States and parts of Europe electric utilities and petroleum refineries tend(ed) to exercise monopoly power at the regional level, pulp and paper production has concentrated in recent decades, and automobile manufacturing is done by a relatively small number of firms in any given country. These examples suggest that non-competitive industries are more than a theoretical curiosity for environmental policy. In response, an enormous literature has arisen that considers the role of market power in all its guises. Researchers have examined output market structures such as monopoly, oligopoly, and monopolistic competition. Homogeneous versus differentiated products, quantity versus price competition, and market power in factor markets have also been considered. In this chapter we limit attention to monopoly and oligopoly output market structures, since these provide useful vehicles for demonstrating the main techniques and conclusions from this larger literature.

The early literature on environmental policy and market power provides additional motivation for studying non-competitive output markets. Papers by Buchanan and Stubblebine (1962) and Buchanan (1969) challenged the Pigouvian point of view outright, by noting that a monopolist distorts the market by producing too little output.

Via the 22 previous chapters in this book we have tried to tell the story of environmental economics, beginning with its roots in neoclassical welfare theory and the environmental movements of the 1960s, and through to the sophisticated theoretical and applied research that today defines the field. We have attempted to integrate in a single volume the conceptual, empirical, and policy aspects of the field, without compromising on the depth of coverage. This, more than anything, explains the length of the book and the amount of time it has taken to write. Even with this, there are many topics that deserve discussion which we have elected not to cover in the book. In this chapter we will briefly discuss some of the important omissions and offer some closing thoughts.

We begin in the next section by noting that environmental economics, like many other fields in microeconomics, has experienced a flowering of empirical research. This is due to several simultaneously occurring factors, including greater appreciation for careful applied work in the economics profession, advances in applied econometric methodology, and the increased availability of high-quality behavioral and environmental data. Many of the questions being pursued do not fit cleanly into the standard canon of the field, which has served as the organizing principle of the book. To highlight this increasingly important aspect of the field, in section 23.1 we summarize a handful of recent papers that illustrate the range of empirical research questions that have occupied environmental economists.

In addition to specifically empirical topics, there are several other areas that we have not covered or have discussed only briefly. Examples include the large economic literature on climate change, topics at the nexus of energy and environmental economics, and political economy/environment issues related to interest group activities, alternative regulator objective functions, and voting. We have also omitted traditional resource economics, and topics that combine elements of environmental policy and natural resource management. In section 23.2 we catalog these topics, discuss their omission in the context of the book's organization and intent, and provide references for further reading.

Our emphasis thus far in Part II of the book has been partial equilibrium, in the sense that we have focused only on the polluting sector in our consideration of environmental policy. When studying competitive markets from this perspective, we have consistently found that the optimal pollution level occurs at the point where firms’ marginal abatement costs are equal to society's marginal damages. Furthermore, the short run performance of the full suite of incentive-based policies was equivalent. Importantly, the extent to which the different policies raise revenue for the government was important only for equity, and did not play a role in efficiency comparisons. Thus emissions could be taxed or abatement subsidized, and pollution permits auctioned or freely distributed, and the only difference was in who captured the environmental scarcity rents. Things were slightly different in the long run, in that abatement subsidies could cause excess entry and were therefore inferior to emissions taxes. Nonetheless, we found that permit schemes that do not raise revenue perform equally to their auctioned counterparts, so long as total emissions are capped when firms enter or leave the market. The ability to separate equity considerations from efficiency outcomes was touted as an important source of flexibility available to the environmental regulator.

A question we have not yet examined is the extent to which these partial equilibrium results continue to hold when we consider how environmental policy might spill over into other markets. We saw in Chapter 5, for example, that environmental policy raises the output price of the polluting good. This can affect the demand for other consumption goods, impact households’ labor supply decisions, and cause changes in real income. More importantly, pollution externalities are usually not the only sources of distortions in an economy. Government revenue is often raised via proportional labor taxes, thereby encouraging the overconsumption of leisure, and causing potentially substantial welfare losses. Other factor input markets are also distorted, as are many product markets through value added taxes or subsidies. It is natural to ask how these pre-existing distortions might interact with environmental policy, causing us to alter the conclusions obtained when such spillover effects are ignored.

Thinking on this topic emerged along two lines. The first focused on the potential for environmental policy to result in a “double dividend”.

In Part II of this book we focus on the design of environmental policy by considering several generalizations and extensions to the model developed in Chapter 3. We begin in Chapter 4 by examining the effects of uncertainty on policy design. Here we will see that the welfare equivalence of emission tax and cap and trade systems, established in Chapter 3, breaks down under some types of uncertainty. In Chapters 5 and 6 we consider how competitive and imperfectly competitive output markets, respectively, interact with policy design. The focus on competitive output markets also enables an initial look at distributional issues in environmental policy, and our analysis of imperfect competition allows us to introduce second-best policy considerations. Second-best considerations then become the primary focus in Chapter 7, in which we examine how environmental policy interacts with other, pre-existing distortions in the economy, such as income taxes. This yields some important insights that differentiate revenue raising and non-revenue raising policy prescriptions. In Chapter 8, we consider a range of topics that are relevant for the real-world design and performance of cap and trade systems. It is here, for example, that we formally address the spatial and temporal aspects of environmental policy, and consider issues related to monitoring and enforcement. In Chapter 9, we introduce the non-point pollution problem, whereby a potentially stochastic ambient pollution level is observed, but emissions cannot be attributed to individual polluters. We will see that the inability to observe polluter behavior creates challenges for implementing the policy options discussed in the earlier chapters, and explains the relative dearth of economic incentive-based approaches to regulating many types of non-point source pollution.

Taken as a whole, Chapters 4 through 9 cover the main topics related to domestic environmental policy for flow pollutants, and they constitute the core themes in many graduate offerings on environmental policy. In the remainder of Part II, we cover several additional, and in some cases more advanced, topics. We begin in Chapter 10 with a discussion of liability rules. When an economic activity creates the risk of an environmental spill, as with nuclear power production or petroleum transport, liability rules can be designed that incentivize the proper level of accident prevention effort. We study the performance of liability rules under different assumptions about risk reduction effort observability, solvency, organizational structure, and insurance coverage.

In our discussion to this point, we have assumed that there is perfect and symmetric information among all agents involved in the design and execution of environmental policy. This has allowed us to refer to damage and cost functions that are fully observed by the regulator, firms, and households alike. Among other things, the full information assumption helped us establish the equivalency of auctioned permits and emission taxes, and more generally suggested that a particular pollution reduction goal can be obtained using one of several economic incentive instruments – all of which have similar efficiency properties. Though it is pedagogically useful, a perfect information assumption is a clear departure from reality. In this chapter we begin to consider how imperfect information may alter the effectiveness and efficiency properties of the policy instruments.

Imperfect information plays a large role in environmental economics, and as reviewed by Pindyck (2007), it comes in many guises. In this chapter we focus on a particular type of uncertainty: the regulator's inability to fully observe aggregate pollution damage and abatement cost functions. Importantly, we assume for this analysis that firms know their own abatement cost functions with certainty. Once we acknowledge that these functions must be estimated by the regulator (inevitably with error), it is natural to ask about the extent to which estimation error influences the performance of different regulatory approaches. Casual intuition suggests estimation errors will result in imprecise policy, but will not cause us to systematically favor one approach over another. Investigation of the extent to which this intuition holds began with Weitzman's (1974) classic paper, and has continued in various forms to this day. The unifying theme in this large literature is a search for policy instruments that minimize the ex post inefficiencies that occur due to the regulator's uncertainty. A related literature beginning with Kwerel (1977) examines ways that the regulator can design mechanisms that resolve the uncertainty, thereby eliminating ex post inefficiencies. In this chapter we examine these two strands of literature.