YALE LAW SCHOOL

Public Law & Legal Theory Research Paper Series Research Paper No. 58

Environmental Protection in the Information Age

by Daniel C. Esty

This paper can be downloaded without charge from the Social Science Research Network Paper Collection at http://papers.ssrn.com/abstract=429580

ENVIRONMENTAL P ROTECTION IN THE INFORMATION AGE Draft July 2003 Forthcoming NYU Law Review, volume 79 (2004)

Daniel C. Esty* Yale Law School P.O. Box 208215 tel: (203) 432-1602/6256 fax: (203) 432-4871/3817

*

Yale Law School and Yale School of Forestry and Environmental Studies; former Deputy Chief of Staff and Deputy Assistant Administrator for Policy at the US Environmental Protection Agency. Thanks to Farrin Anello, Mark Barnett, Noah Chesnin, Katy Fischer, Brian Fletcher, Phil Fortino, Diane Haar, Dean Kawamoto, Julia Peck, and Meryl Raymar for research assistance. Thanks to Bishop Grewell, David Victor, Nicole Vickey, Thiru Vignarajah, David Roe, Dennis Hirsch, and Rob Klee as well as the participants in the Yale Law School faculty seminar, the Yale Environment School faculty seminar, and INSEAD’s environmental seminar for comments on prior drafts.

ENVIRONMENTAL PROTECTION IN THE INFORMATION AGE TABLE OF CONTENTS I.

INTRODUCTION ................................................................................................................................1

II.

INFORMATION IN THE ENVIRONMENTAL R EALM...........................................................................5 A. Information Issues in the Academic Literature ..............................................................8 B. Information-Issues in Market and Regulatory Failures................................................13 1. Property versus Liability Rules.........................................................................15 2. Beneath the Cathedral.......................................................................................16 a. Problem Identification...........................................................................17 b. Causal Specification..............................................................................18 c. Impact Evaluation .................................................................................19 d. Harm Valuation.....................................................................................20 e. Rights Delineation.................................................................................21 f. Policy Intervention.................................................................................22 g. Implementation .....................................................................................23 h. Monitoring and Enforcement ................................................................23 i. Updating and Refinement ......................................................................24 III.

GAPFILLING AND INSTITUTIONAL D ESIGN .......................................................................26 A. Obstacles to Gapfilling.................................................................................................28 1. Technical Deficiencies ......................................................................................29 2. Cognitive Shortcomings....................................................................................30 3. Structural Mismatches.......................................................................................32 4. Public Choice Distortions .................................................................................33 5. Administrative Inefficiency ..............................................................................34 B. Collectivization v. Decentralization............................................................................34 C. Optimal Specificity of Regulation................................................................................37 D. Corrective Justice .........................................................................................................40

IV.

INFORMATION AGE OPPORTUNITIES FOR ENVIRONMENTAL PROGRESS .......................44 A. Bringing the Information Age to Bear on Environmental Protection .................................................................................44 1. Computer Power ...............................................................................................44 2. Communications ...............................................................................................45 3. Internet ..............................................................................................................47 4. Digitalization.....................................................................................................49

V.

RETHINKING ENVIRONMENTAL PROTECTION IN AN INFORMATION-R ICH WORLD.......50 A. Revitalizing the Environmental Property Rights Market.............................................51 1. Search Costs ......................................................................................................51 2. Scope of Environmental Rights ........................................................................52 3. Property Demarcation.......................................................................................54 4. Valuation and Strategic Behavior .....................................................................56 5. De-Collectivization...........................................................................................58

B. C.

D. E. F.

6. Contracting Beyond Small Numbers ................................................................58 Court Protected Entitlements .......................................................................................59 Command and Control Regulation...............................................................................60 1. Technical Deficiencies ......................................................................................61 2. Cognitive Shortcomings....................................................................................63 3. Structural Mismatches.......................................................................................64 4. Public Choice Distortions .................................................................................65 5. Administrative Inefficiency ..............................................................................68 Market-Based Regulation.............................................................................................70 Social Context and New Modes of Collective Action..................................................72 Institutional Realignment ..............................................................................................74 1. Decollectivization and the Rise of Market Approaches ...................................74 2. The End of “Do Nothing” .................................................................................76 a. No Visible Harm Means No Harm........................................................76 b. Myth of Endless Abund ance .................................................................77 c. Assumption of Rough Reciprocity (“live and let live”) ........................78 d. Small Equals Zero .................................................................................79

VI. DRIVING ENVIRONMENTAL INFORMATION INTO THE D ECISIONMAKING R EALM ..............80 A. Beyond Externalities ....................................................................................................81 B. Environmental Information as a Public Good ..............................................................83 1. Critical Environmental Information Does Not Exist ........................................84 a. Technical and Cost Barriers ..................................................................84 b. Market Failures .....................................................................................87 c. Legal and Institutional Failures.............................................................89 2. Information Withheld ........................................................................................94 a. Antagonism Toward Government .........................................................94 b. Competitive Positioning........................................................................95 VII.

DOWNSIDES OF THE INFORMATION AGE ..........................................................................96 A. New Environmental Harms..........................................................................................96 B. Negative Impacts on Environmental Decisionmaking.................................................98 1. Will Information Become Usable Knowledge? ................................................98 2. Information as a Strategic Asset .....................................................................100 3. Network Effects ..............................................................................................100 4. Transparency Traumas ....................................................................................101 5. Downsides of Cyberdemocracy......................................................................102

VIII. CONCLUSION ....................................................................................................................103

I. Introduction At the height of the dotcom boom, Digital Age gurus extolled the revolutionary power and benefits of emerging information technologies and confidently declared that the “Internet changes everything.”1 These voices are quieter now. Indeed, many have been silenced by bankruptcy. But were they wrong? In this Article, I argue that the emerging technologies of the Information Age — computers, the Internet, remote sensing, and so on — will be transformative, and I explore how these breakthroughs will affect both the theory and practice of environmental law. Information issues lie at the core of the challenge of environmental protection. Inadequate information and related transaction costs make a Coasian world of contractual exchanges of environmental rights unfeasible in most circumstances.2 Information issues also lie at the heart of the debate over whether entitlements, including environmental rights, are best protected by property rules or liability rules.3 But even as a generation of legal scholars debated the optimal form of tort law, mounting frustration over the later half of the 20th century with the information demands of a tort-based environmental regime precipitated a shift toward a structure based on statutes and regulations.4 In recent

1

See, e.g ., Desiree J. Hanford, Unwise Wisdom: The Internet Changes Everything, W ALL ST . J., Jan. 29, 2001, at R12 (discussing how the Internet has changed capital markets by slashing transaction costs); Alan Frutkin, Spinning the Web, M EDIAW EEK, Jan. 24, 2000, at 52 (examining the impact of the Internet on television and popular culture); Jay Hoag, Heard from the Buy Side, RED HERRING (August 1999), available at http://www.redherring.com/mag/issue69/invbuyside.html (describing how the Internet has transformed Wall Street); Dan Johnson, The Internet Alters Politics, FUTURIST , Jan./Feb. 2000, at 11 (reporting on the impact of the Internet on the public policy formation activities of journalists, legislators, and lobbyists); DEMOCRACY.COM: GOVERNANCE IN A NETWORKED W ORLD (Elaine C. Kamarck & Joseph S. Nye, Jr., eds., 1999) (exploring the ways in which the Internet is revolutionizing governmental institutions and decisionmaking). 2

R. H. Coase, The Problem of Social Cost, 3 J. L. & ECON. 1 (1960) (arguing that, in the absence of transaction costs, an efficient level of emissions can be negotiated) [hereinafter Coase, Social Cost]; COASEAN ECONOMICS: LAW AND ECONOMICS AND THE NEW INSTITUTIONAL ECONOMICS (Steven G. Medema ed., 1997); Robert C. Ellickson, The Case for Coase and Against “Coaseanism”, 99 YALE L. J. 611 (1989) [hereinafter Ellickson, Coaseanism]; Guido Calabresi, The Pointlessness of Pareto: Carrying Coase Further, 100 YALE L. J. 1211 (1991) [hereinafter Calabresi, Pointlessness]. 3

See Guido Calabresi & Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View of the Cathedral, 85 HARV. L. REV. 1089 (1972) (arguing that efficiency requires that costs be put on party best able to make the costbenefit calculation) [hereinafter Calabresi & Melamed, Cathedral]. For the broader Cathedral literature, see fn. ___ infra. 4

James E. Krier & Steward J. Schwab, Property Rules and Liability Rules: The Cathedral in Another Light, 70 N.Y.U. L. Rev. 440 (1995) (noting that problems in obtaining and processing information may impede efficient damage calculations by the courts) [hereinafter Krier & Schwab, Property Rules]; see also Richard B. Stewart, A New Generation of Environmental Regulation? 29 CAP . U. L. REV . 21 (2001) [hereinafter Stewart , A New Generation] (chronicling and criticizing the rise of the environmental regulation model); and ROBERT V. PERCIVAL, ENVIRONMENTAL REGULATION: LAW

years, similar disappointment has arisen over regulatory failures — often traceable to information gaps — that remain pervasive despite numerous regulatory reform initiatives.5 Uncertainty seems to be the hallmark of the environmental domain. Disagreements over how best to cope with information deficits have translated into bitterly partisan and divisive environmental politics and limited progress in recent years in pollution control and natural resource stewardship. Perhaps this picture represents the inescapable reality of the environmental realm. But imagine instead another world, one where pollution (or the use of natural resources6 ) could be easily traced at low cost. What if each increment of SO x or NOx emitted from a smokestack could be tracked to where it lands downwind? And what if the “fate and transport” (as an ecological scientist would put it) of nitrogen and phosphorus in runoff from farm fields or suburban lawns could easily be determined? Would things not be different if automobile emissions of volatile organic compounds (VOCs), particulates, and other pollutants could be tagged and followed as they flow from tailpipes to the “receptors” that absorb them? And what if the impacts on the receptors (including people, crops, lakes, and forests) were better understood as a result of advances in epidemiology, risk analysis, and various ecological sciences? Better data on the scope, spread, and effects of environmental harms would not address all of the gaps that plague environmental policymaking. But such information would make internalizing many externalities much easier and would alter the framework within which we think about environmental problems. Such a vision is not really so farfetched. Computers, modern telecommunications, sensor technologies, nanotechnologies, and other Information Age breakthroughs are reshaping every facet of modern life. And although their impact is just beginning to be felt in the environmental realm, the

SCIENCE AND POLICY 464 (2d ed. 1996) (providing a comprehensive structural overview of the emergence and eventual dominance of statutory environmental law). 5

J. CLARENCE DAVIES & JAN M AZUREK, POLLUTION CONTROL IN THE UNITED STATES : EVALUATING THE SYSTEM (1998) [hereinafter DAVIES & M AZUREK, POLLUTION CONTROL]. 6

The environmental realm encompasses a broad array of pollution and natural resource management issues. This Article addresses this spectrum to some degree but places primary focus on pollution control.

emergence of powerful new digital technologies and their deployment across the economy and society is now broadly evident and inexorable.7 Because information gaps and uncertainties lie at the heart of many persistent pollution and natural resource management problems, these breakthroughs offer special promise in the environmental domain. As data become more plentiful, easier to move and manipulate, and dramatically less costly, 8 our capacity to spot problems, understand environmental threats, place values on harms, manage regulatory complexity, evaluate policy options, tighten feedback loops, and refine environmental policies will increase substantially. In such an information-rich world, both the environmental problem set and our response strategies seem likely to be reconfigured. The emergence of a wide-ranging set of pollution tracking, taxing, and trading mechanisms creates the potential for new mechanisms to internalize externalities and limit environmental harm. For example, information generated by satellite-based remote sensing devices can be deployed as “virtual barbed wire,” making it easier to delineate and defend property rights such as tradable fishing quotas. A more robust property rights regime makes a Coasean World of contractual exchange of environmental rights potentially functional in more cases. The same technological advances which make measuring and monitoring emissions much easier will also permit a shift toward greater use of market-based regulatory approaches such as pollution taxes and emissions allowance trading. Even where traditional command-and-control approaches to regulation persist, the zone of uncertainty surrounding decisions will be reduced9 and foundations for policymaking will be improved, potentially reducing the degree of regulatory failure. In addition, the advances of the Information Age promise to speed up the process of norm diffusion and change the ways that societal preferences are “rolled up,” 7

PETER SCHWARTZ, PETER LEYDEN, & JOEL HYATT, THE LONG BOOM 26 (1999) (discussing the “inevitable” change and growth being driven by the spread of computer technology) 8

BILL GATES , THE ROAD A HEAD (1995) [hereinafter GATES , THE ROAD A HEAD] (identifying a range of emerging breakthrough technologies); W ILLIAM J. M ITCHELL, E-TOPIA 13 (1999) [hereinafter M ITCHELL, E-TOPIA] (“[I]nformation has become dematerialized and disembodied; it is now whizzing round the world at warp speed, and in catex-crackling quantities”); Ira S. Nathanson, Internet Infloglut and Invisible Data: Spamdexing Search Engines with Meta Tags, 12 HARV. J. L. & TECH. 43, 51 (1998) (describing the explosion of information accessible through free online search engines). 9

Daniel C. Esty, Toward Optimal Environmental Governance, 74 N.Y.U. L. REV . 1495, 1519 (1999) [hereinafter Esty, Environmental Governance].

which may create a social context of greater environmental concern that translates into more voluntary action to control pollution and promote natural resource stewardship. Downsides to the application of Digital Age technologies in the environmental realm must also be anticipated. Not all information advances translate into knowledge gains. Lower cost information access may mean more disinformation circulates through the policymaking process. And the shift towards a digitized world may not facilitate better decisionmaking and citizen involvement in environmental protection efforts. Instead it may increase the opportunities for special interest manipulation of the policy process as well as the risk of public distraction and citizen disengagement from policymaking. This Article seeks to look forward and to understand how the Information Age might realign our environmental protection regime. The effort is both descriptive and normative. It proceeds in several stages. Part II surveys the existing “information and environment” literature and concludes that the wide range of ways that data gaps and knowledge deficiencies affect not only our choice of rules and policies, but also our institutions and strategies for addressing environmental harms, has been underappreciated. In mapping this terrain, I suggest that addressing information gaps represents a key if not the key to environmental progress. In Part III, I discuss the “gapfilling” that is required to make good environmental decisions. Building on Oliver Williamson’s institutional focus on markets and hierarchies, I analyze the set of decisionmaking institutions available in the environmental domain and the ways that the gapfilling process might break down. I argue that information gaps and the cost of filling them shapes the ends as well as the means of environmental protection. I survey in Part IV the key Information Age developments that promise to reshape environmental law and policymaking. Some of the emergent technologies, such as computers, are already in broadbased use. Others, such as nanotechnologies are just beginning to be seen and understood. In almost all cases, however, the application of these breakthroughs to environmental challenges has not been closely examined.

In Part V, I look across our current institutional approach to environmental protection and analyze how a more information-rich world could affect existing policy tools and strategies. I suggest that information technologies could enhance many of our present approaches to pollution control and natural resource management. And I observe that the optimal mix of environmental strategies is likely to change as these technologies permeate modern. Although the gains from a more information-rich world will in many cases penetrate the environmental regime on their own, Part VI explores how this process might be enhanced. I review who might be motivated to generate needed information and what could be done to get the information to decisionmakers. Starting with a focus on the “le ast-cost information providers,” I identify a set of options to restructure the incentives for data production and analysis in pursuit of a more analytically rigorous environmental policy process. Part VII acknowledges that Information Age breakthroughs may create new problems in the environmental domain and may exacerbate existing policymaking pathologies. Thus, I develop a taxonomy of potential downsides from the emergence of digital technologies. This analysis provides a “watch list” of issues that mig ht require policy intervention in the next few years. Part VIII provides some preliminary conclusions on what we might expect from an era of Information Age environmental protection. I see considerable opportunity for improved pollution control and better natural resource management. But I caution that the precise pace and path of change is hard to predict and may result in steps backward as well as forward.

II.

Information in the Environmental Realm Information gaps plague environmental decisionmaking from the household level10 to the global

10

Christine Jolls et al., A Behavioral Approach to Law and Economics, in BEHAVIORAL LAW AND ECONOMICS 42-45, 47 (Cass R. Sunstein ed., 2001) (noting consumer difficulties in getting and processing risk information) [hereinafter Jolls et al., A Behavioral Approach]; W ESLEY A. M AGAT & W. KIP VISCUSI, INFORMATIONAL APPROACHES TO REGULATION (1992) (highlighting risk detection problems individuals face).

scale.11 The U.S. Environmental Protection Agency has repeatedly acknowledged significant shortcomings in the core data sets on which regulatory decisions depend. 12 Many other observers of the policymaking process have also stressed that information problems represent a fundamental issue holding society back from better results in pollution control and natural resource management.13 As a theoretical matter, the importance of good data and solid analytic information to sound environmental decisionmaking seems obvious. Without basic data, externalities cannot be identified and internalized. And the victims of pollution who bear the cost of emissions harms cannot be made whole. It has thus long been understood that information gaps lead to market failures,14 legal system breakdowns,15 and regulatory difficulties.16

11

INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE, CLIMATE CHANGE 2001: IMPACTS, A DAPTATION, AND VULNERABILITY – SUMMARY FOR POLICYMAKERS 17 (2001) (identifying high priority areas for narrowing gaps between current knowledge and policymaking needs); Daniel Bodansky, The United Nations Framework Convention on Climate Change: A Commentary, 18 YALE J. INT ’L L. 451, 455-57, 475-78 (1993) (giving overview of information issues related to climate change). 12

See, e.g ., EPA, NATIONAL W ATER QUALITY INVENTORY (1996) (revealing serious water data limitations); EPA, REDUCING RISK: SETTING PRIORITIES AND STRATEGIES FOR ENVIRONMENTAL PROTECTION (1990) (identifying numerous data gaps). 13

See, e.g ., Karl Hausker, Reinventing Environmental Regulation: The Only Path to a Sustainable Future, 29 ENVTL. L. REP . 10148 (1999); STEPHEN BREYER, BREAKING THE VICIOUS CIRCLE (1993) [hereinafter BREYER, BREAKING]; GAO, M AJOR M ANAGEMENT CHALLENGES AND PROGRAM RISKS : ENVIRONMENTAL PROTECTION A GENCY 14-21 (GAO-01-257 2001); GAO, ENVIRONMENTAL INFORMATION: EPA NEEDS BETTER INFORMATION TO M ANAGE RISKS AND M EASURE RESULTS (GAO-01-97T, Oct. 3, 2000). 14

The “economics of information” and “search costs” have been a critical point of analysis in the economic realm. J. Wallis & D. North, Measuring the Transaction Sector in the United States Economy, 1870-1970, in LONG TERM FACTORS IN A MERICAN ECONOMIC GROWTH (S. Engerman & R. Gallman eds., 1986); Steve Salop, Information and Monopolistic Competition, 66 A MER. ECON. REV. 240, 240-44 (1976); George Akerlof, The Market for “Lemons”: Qualitative Uncertainty and the Market Mechanism, 84(3) QUART . J. ECON. 488-500 (1970); George J. Stigler, The Economics of Information, 69 J. POL. ECON. 213 (1961); Joseph E. Stiglitz, The Contributions of the Economics of Information to Twentieth Century Economics, 115 Q. J. ECON. 1441 (2000) (discussing information asymmetries and costs and their implications for markets). 15

A. Mitchell Polinsky, Resolving Nuisance Disputes: The Simple Economics of Injunctive and Damage Remedies, 32 STAN. L. REV. 1075 (1980) (discussing difficulties courts have in making proper assessments); PETER H. SCHUCK, THE LIMITS OF LAW 438-40 (2000) [hereinafter SCHUCK, LIMITS OF LAW ] (arguing that the court system is often overwhelmed by requirements) Krier & Schwab, Property Rules, supra note 4 (highlighting failures in the tort regime). 16

NEIL K. KOMESAR , IMPERFECT A LTERNATIVES : CHOOSING INSTITUTIONS IN LAW , ECONOMICS, AND PUBLIC POLICY (1994) (the economy of information determines how people use information); DAVIES & M AZUREK, POLLUTION CONTROL, supra note 5; M ARC LANDY, ET AL., THE ENVIRONMENTAL PROTECTION A GENCY: ASKING THE WRONG QUESTIONS (1992) [hereinafter M ARC LANDY, ET AL., A SKING THE W RONG QUESTIONS ].

But much of the academic literature on how to improve environmental performance has touched on information issues only indirectly or in passing. Significant scholarly work has been done on many sources of regulatory failure including public choice distortions in policymaking,17 federalism and questions concerning which level of government should undertake environmental responsibilities,18 and risk perception problems. 19 Additional emphasis has been placed on regulatory inefficiency and strategies—such as economic incentives,20 “regulatory competition,” 21 and improved cost-benefit analysis 22 —for addressing bureaucratic shortcomings. While data gaps are sometimes mentioned as an issue, pervasive uncertainties are simply assumed by most scholars to be part of the framework within which environmental law must operate.23

17

See generally JERRY L. M ASHAW , GREED, CHAOS , AND GOVERNANCE: USING PUBLIC CHOICE TO IMPROVE PUBLIC LAW (1997) [hereinafter M ASHAW , GREED]; Matthew D. McCubbins et al., Administrative Procedures as Instruments of Political Control, 3 J. L. ECON. & ORG. 243 (1987). 18

See generally Richard B. Stewart, Pyramids of Sacrifice? Problems of Federalism in Mandating State Implementation of National Environmental Policy, 86 YALE L.J. 1196 (1977). 19

See generally BREYER, BREAKING, supra note 13; JOHN D. GRAHAM & JONATHAN B. W IENER, RISK V. RISK (1995) [hereinafter GRAHAM & W IENER, RISK V. RISK]; Cass Sunstein & Timur Kuran, Availability Cascades and Risk Regulation, 51 STAN. L. REV. 683 (1999) [hereinafter Sunstein & Kuran, Availability Cascades]; PAUL SLOVIC , THE PERCEPTION OF RISK (2001) [hereinafter SLOVIC , PERCEPTION OF RISK]; Jeffrey J. Rachlinski & Cynthia R. Farina, Cognitive Psychology and Optimal Government Design, 87 CORNELL L. REV . 549 (2002) [hereinafter Rachlinski & Farina, Cognitive Psychology] (surveying cognitive sources of regulatory failure). 20

Bruce Ackerman & Richard Stewart, Reforming Environmental Law : The Democratic Case for Market Incentives, 13 COL. J. ENVTL. L. 171 (1988); Robert Hahn & Robert Stavins, Incentive-Based Environmental Regulation: A New Era from an Old Idea? 18 ECOLOGY L. Q. 1 (1991); PROJECT 88—HARNESSING M ARKET FORCES TO PROTECT THE ENVIRONMENT : INITIATIVES FOR THE NEW PRESIDENT (Robert Stavins ed., 1988). 21

Charles Tiebout, A Pure Theory of Local Expenditures, 63 J. POL’Y. & ECON. 416 (1956); Richard L. Revesz, Rehabilitating Interstate Competition: Rethinking the “Race to the Bottom” Rationale for Federal Environmental Regulation, 67 N.Y.U. L. REV . 1210 (1992); Daniel C. Esty, Revitalizing Environmental Federalism, 95 M ICH. L. REV. 570 (1996) [hereinafter Esty, Revitalizing].See generally REGULATORY COMPETITION AND ECONOMIC INTEGRATION: COMPARATIVE PERSPECTIVES (Daniel C. Esty & Damien Geradin eds., 2001) [hereinafter REGULATORY COMPETITION]. 22

W. KIP VISCUSI, FATAL TRADEOFFS: PUBLIC AND PRIVATE RESPONSIBILITIES FOR RISK (1992); Corrosion Proof Fittings v. EPA, 947 F.2d 1201 12222-24 (5th Cir. 1991). BREYER, BREAKING, supra note 13. 23

ROBERT V. PERCIVAL ET AL., ENVIRONMENTAL REGULATION : LAW , SCIENCE, AND POLICY 387 (3d ed. 2000) (emphasizing the pervasive uncertainty associated with environmental policy in the context of risk assessment) [hereinafter PERCIVAL ET AL., ENVIRONMENTAL REGULATION]; Bradley C. Karkkainen, Information As Environmental Regulation: TRI and Performance Benchmarking, Precursor To A New Paradigm, 89 GEO. L. J. 257 (2001) [hereinafter Karkkainen, Benchmarking] (emphasizing that the world of environmental regulation is characterized by uncertainty and data gaps); Calabresi, Pointlessness, supra note 2, at 1218 (information gaps are pervasive and thus a “degree of friction is a reality of life”).

A. Information Issues in the Academic Literature There are a few exceptions to this pattern. A line of scholarship has emerged in recent years around the potential for “information regulation.”24 Much of this analysis focuses on EPA’s Toxic Release Inventory (TRI) and the disclosure of environmental information as a means for changing corporate behavior.25 Pederson argues that the value of disclosure programs goes beyond affecting the incentives facing the business community to the possibility of provoking a public outcry that will result in new regulation. 26 Sunstein similarly sees value in information disclosure as a potential regulatory tool which triggers political rather than market safeguards.27 Other scholars have emphasized the value of providing information to citizens in their role as consumers.28 Whether aiming to shame polluters into behaving better, inspire regulators, or engage consumers, those writing in this vein all seek to tweak the policy structure through revised information flows. But they fail to take up the deeper role of information—and especially information gaps—as a theoretic al matter and as a fundamental source of both market and regulatory failures.

24

A NTHONY I. OGUS , REGULATION: LEGAL FORM AND ECONOMIC THEORY 121-49 (1994) (explaining the concept and providing examples of information regulation); see also David W. Case, The Law and Economics of Environmental Information as Regulation, 31 E. L. R. 10773 (2001) (discussing the economic underpinnings of information regulation). 25

James Hamilton, Pollution is News: Media and Stockmarket Reactions to the Toxics Release Inventory Data, 28 J. ENVTL. ECON. & M GMT . 98 (1995) (looking at the connection between TRI data and corporate stock prices); Shameek Konar & Mark Cohen, Information as Regulation: The Effect of Community Right to Know Law on Toxic Emissions, 32 J. ENVTL. ECON. & M GMT . 109 (1997) (a further examination of the relationship between TRI releases and stock prices); Tom Tietenberg, Disclosure Strategies for Pollution Control, 11 ENVTL. & RESOURCE ECON. 587 (1998) (evaluating the effectiveness of various information disclosure programs); Karkkainen, Benchmarking, supra note 23 (evaluating the effectiveness of the TRI program). 26

William Pederson, Regulation and Information Disclosure: Parallel Universes and Beyond, 25 HARV. ENVTL. L. REV. 151 (2001). 27

Cass R. Sunstein, Informational Regulation and Informational Standing: Akins and Beyond, 147 U. PENN. L. REV. 613 (1999) [hereinafter Sunstein, Informational Regulation]. But note Sunstein also suggests that disclosure strategies may be ineffectual or more costly than traditional command and control approaches. 28

Stephen Breyer, Analyzing Regulatory Failure: Mismatches, Less Restrictive Alternatives and Reform, 92 HARV. L. REV. 549, 556 (1979); Peter S. Menell, Structuring a Market-Oriented Federal Eco-Information Policy, 54 M D. L. REV. 1435 (1995) (evaluating eco-labels and other programs aimed at informing consumers); James Salzman, Informing the Green Consumer: The Debate Over the Use and Abuse of Environmental Labels, 1 J. IND. ECOL. 11 (1997) (discussing ecolabels).

A related line of scholarship focuses on the possibility that more information, systematic analysis of data, and better environmental management systems will make corporations more “reflexive” and thus attentive to the demands of environmental stewardship. 29 Others have looked at the role of information in specific regulated communities. Lyndon, for instance, argues that laws addressing chemical exposures should be re-crafted to encourage production and dissemination of toxicity information. 30 Wagner sees similar gains in restructuring the incentives for information generation related to toxic products.31 McGarity and Shapiro suggest that public disclosure of test results involving pesticides could enhance environmental safety.32 Sage argues that mandatory disclosure laws have value insofar as information is a public good. 33 And Florini envisions a new world of “regulation by revelation.”34 A broader line of information-related scholarship has been developed under the rubric of “industrial ecology.” One of the central tenets of this emerging field is that the flows of material and energy across the economy must be traced with greater precision if we are to understand environmental problems systemically. 35 Better information thus emerges as a key tool for improving efficiency and reducing emissions in the industrial sector, facilitating closed loop production processes, 29

Eric W. Orts, Reflexive Environmental Law, 89 NW . U. L. REV. 1227, 1252-57 (1995) (developing the theory of reflexive environmental law); Stewart, A New Generation, supra note 4 (discussing information regulation, including environmental audit and environmental management systems, as a way to drive a more reflexive approach to environmental law). 30

Mary L. Lyndon, Information Economics and Chemical Toxicity: Designing Laws to Produce and Use Data, 87 M ICH. L. REV. 1795 (1989) [hereinafter Lyndon, Information Economics]. 31

Wendy Wagner, Choosing Ignorance in the Manufacture of Toxic Products, 82 CORNELL L. REV. 773 (1997) [hereinafter Wagner, Choosing Ignorance] (spelling out how better toxics control might be possible if incentives were in place to generate information). 32

Thomas O. McGarity & Sidney A. Shapiro, The Trade Secret Status of Health and Safety Testing Information: Reforming Agency Disclosure Policies, 93 HARV. L. REV . 837 (1980). 33

William M. Sage, Regulating Through Information: Disclosure Laws and American Health Care, 99 COLUM. L. REV. 1701 (1999). 34

A NN FLORINI, THE COMING DEMOCRACY: NEW RULES FOR RUNNING A NEW W ORLD (2003) [hereinafter FLORINI, COMING DEMOCRACY] (discussing disclosure and transparency as the emerging centerpiece for regulation). 35

Thomas E. Graedel, On the Concept of Industrial Ecology, 21 A NN. REV. ENERGY & ENV’T . 69 (1996) (explaining the concept of industrial ecology and noting the connections to greater environmental efficiency); Robert J. Klee, Enabling Environmental Sustainability in the United States: The Case for Material Flow Inventory, 23 STANFORD ENVTL. L. J. (forthcoming 2004) (arguing that compiling comprehensive information on society’s use of materials and energy is a key first step towards ultimate environmental sustainability).

“dematerialization,” and more careful “design for the environment.”36 Industrial ecologists have furthermore begun to explore the implications of improved data management, and therefore greater environmental efficiency, up and down the supply chain. 37 And some work has begun on the “ecology” of the Information Age Economy.38 Those who seek to re-orient environmental protection around property rights and a common law legal strategy for protecting these rights have also looked at the role of information. 39 But in their zeal to promote a private-property-rights-oriented approach to environmental protection, these advocates for “common law” environmentalism often fail to appreciate the reality of market failures.40 Others take more seriously the problem of information gaps and seek to understand when and why property rights regimes break down. Yandle, for instance, talks about the need for technologies and institutions that can help to make property rights definable, defendable, and divestible.41 36

Braden R. Allenby, The Information Revolution and Sustainability: Mutually Reinforcing Dimensions of the Human Future, in INFORMATION SYSTEMS AND THE ENVIRONMENT (Deanna J. Richards et al. eds., 2001) (demonstrating a “fundamental co-evolution of the information revolution and sustainability”); Thomas E. Graedel & Robert J. Klee, Urban and Anthroposystem Metabolism, in 3 THE ENCYCLOPEDIA OF GLOBAL ENVIRONMENTAL CHANGE 73 (T. Munn ed., 2001) (discussing how industrial metabolism studies of the flow of materials and energy are important diagnostic tools for environmental improvement). 37

Paul R. Kleindorfer & Eli M. Snir, Environmental Information in Supply-Chain Design and Coordination, in INFORMATION SYSTEMS AND THE ENVIRONMENT (A. U. Richards et al. eds., 2001) (demonstrating how better information can improve environmental performance through minimization of the use of raw materials, transportation optimization, reduced emissions, better logistics, and other means); Daniel C. Esty & Michael E. Porter, Industrial Ecology and Competitiveness, 2 J. IND. ECOL. 35 (1998) (discussing how improved information flows up and down the supply chain may translate into greater resource productivity) [hereinafter Esty & Porter, Industrial Ecology]. 38

See, e.g ., ECOLOGY OF THE NEW ECONOMY: SUSTAINABLE TRANSFORMATION OF GLOBAL INFORMATION COMMUNICATION AND ELECTRONICS INDUSTRIES (Jacob Park & Nigel Roone eds. 2002); TECHNOLOGICAL CHANGE AND THE ENVIRONMENT (Arnulf Grubler et al., eds. 2002) (highlighting environmental effects from technological advances, including digitalization); LESTER R. BROWN, ECO-ECONOMY: BUILDING AN ECONOMY FOR THE EARTH (2001); DIGITAL FUTURES : LIVING IN A NETWORKED W ORLD (James Wilsdon (ed.), 2001); JAN M AZUREK, M AKING M ICROCHIPS: POLICY, RESTRUCTURING, AND GLOBALIZATIO IN THE SEMICONDUCTOR INDUSTRY (1999). 39

See generally TERRY L. A NDERSON & DONALD R. LEAL , FREE M ARKET ENVIRONMENTALISM (1991); BRUCE YANDLE , COMMON SENSE AND COMMON LAW FOR THE ENVIRONMENT : CREATING W EALTH IN HUMMINGBIRD ECONOMIES (1997). 40

See, e.g., Terry L. Anderson, Viewing Wildlife Through Coase-Colored Glasses, in W HO OWNS THE ENVIRONMENT ? (Peter J. Hill & Roger E. Meiners eds., 1998) (advancing a call for contract-based protection of wildlife); see also Todd J. Zywicki, Environmental Externalities and Political Externalities: The Political Economy of Environmental Regulation and Reform, 73 TUL. L. REV. 845 (1999) (arguing that the political externalities of environmental regulation are a more severe problem than the underlying pollution spillovers that the regulation is meant to solve). 41

Bruce Yandle, Legal Foundations for Evolving Property Rights Technologies, in THE TECHNOLOGY OF PROPERTY RIGHTS 2 (Terry L. Anderson & Peter J. Hill eds., 2001).

A small number of legal scholars have begun to look at the underlying role of information in environmental problem solving. In a recent essay, Hirsch highlights the sweeping potential environmental impacts of information technologies and the “new economy.”42 In a thoughtful article analyzing the shift of the economy from manufacturing to services, Salzman briefly explores the potential impacts of the Information Revolution on the environment.43 Although he finds positive environmental feedback loops among the key trends—deindustrialization, information technology gains, economic consolidation, and globalization—Salzman concludes that the decline in manufacturing has been overstated and that the potential for real improvements from the “New Economy” still lie in the future.44 Perhaps the most theoretically significant examination of the role of information in the environmental realm dates back to the work of Krier in the 1970s. Building on the economic analysis of Demsetz and Coase and the “law and economics” work of Calabresi, Krier dissects the information cost implications of various institutional approaches to environmental protection. 45 He suggests (with Montgomery) that externalities arise whenever transaction costs outweigh the gains to be achieved by internalizing spillover harms.46 Krier and Montgomery argue that private property will reduce transaction costs, particularly information needs, in some circumstances, making more efficient resource allocations possible.47 But they note that a private property system entails administrative costs which, in other

42

Dennis D. Hirsch, Globalization, Information Technology, and Environmental Regulation: An Initial Inquiry, 20 VA. ENVTL. L. J. 57 (2001) (offering predictions about the impacts of globalization and the technology revolution on the environmental domain). 43

James Salzman, Beyond the Smokestack: Environmental Protection in the Service Economy, 47 UCLA L. REV. 411 (1999) (arguing that the services economy presents different challenges that must be addressed directly and a point of leverage on traditional manufacturing) [hereinafter Salzman, Beyond the Smokestack ]. 44

Id.; see also Bruce Guile & Jared Cohon, Sorting Out a Service-Based Economy, in THINKING ECOLOGICALLY : THE NEXT GENERATION OF ENVIRONMENTAL POLICY (Marian R. Chertow & Daniel C. Esty eds., 1997) (highlighting the need to restructure environmental policy and law to address the shift in the economy from manufacturing to serv ices) [hereinafter THINKING ECOLOGICALLY]. 45

James E. Krier & W. David Montgomery, Resource Allocation, Information Cost and the Form of Government Intervention, 13 NAT . RES . J. 89, 92-98 (1973). 46

Id. at 93.

47

Id. at 93.

circumstances, may not be justified. 48 Describing regulation as the opposite pole from a private property rights regime, Krier and Montgomery observe that, in some cases, government intervention will minimize information burdens and thus the cost of achieving a more efficie nt resource allocation. Specifically, where understanding a problem, communicating knowledge about the problem, and policing costs are high, a regulatory approach may be superior to a regime of private property rights. 49 In such cases, regulations economize on the particularized information that would otherwise be required. Krier and Montgomery identify two intermediate options that lie between a private property approach and full scale government regulation: subsidization and pricing. 50 These approaches overcome the inherent inefficiency of uniform standards and permit emission limitations to vary with the marginal cost of control for each source.51 Such pricing approaches, which we now call “market-based” regulation, avoid the inefficiency of uniformity and the administrative burden that would be associated with rules tailored to each individual pollution source. Krier’s work, spotting information issues as the crux of the environmental protection challenge, provides an important framework for this Artic le. His argument (with Ursin) that regulation should be understood as an information accrual process based on collective trial and error cuts to the environmental chase.52 Other legal scholars have explored similar issues outside the environmental context. Schuck discusses the information demands of legal systems, markets, and social norms and develops a theory of

48

Id. at 94.

49

Id. at 96.

50

Id. at 97.

51

JAMES KRIER, ENVIRONMENTAL LAW AND POLICY 300-301 (1971); J. H. DALES , POLLUTION, PROPERTY, AND PRICES 85 (1968) (spelling out the economic logic of marginal costs emissions pricing). 52

JAMES KRIER & E. URSIN, POLLUTION AND POLICY: A CASE ESSAY ON CALIFORNIA AND FEDERAL EXPERIENCE WITH M OTOR VEHICLE POLLUTION 1940-1975 (1977) (discussing regulations as an information-centered system of collective trial and error) [hereinafter POLLUTION AND POLICY].

the “economy of information.”53 In a series of articles, Sunstein explores the connections between information and efficient regulation. 54 This Article seeks to extend this line of scholarship. B. Information Issues in Market and Regulatory Failures If the property rights over environmental resources are clearly defined and the transactions costs of negotiating their purchase and sale are negligible, private party “Coasean” bargaining will reallocate entitlements so as to generate an economically efficient outcome which internalizes externalities.55 We do not, however, live in a frictionless world. 56 As we shift from simple two-person models to a multiperson reality it becomes more likely that a contractual approach to environmental protection will fail as information gaps—over who holds the property rights, how much infringement has occurred, what the value of the loss has been, etc.—become too numerous and complicated for the parties to bridge on their own. If the gaps to be filled are relatively narrow and turn largely on who holds the relevant rights, damage assessments, or certain other “legal” issues, tort law (supported by a court system) may be able to fill the breach.57 Where the gaps are broader and involve scientific or technical issues, other institutions, notably regulatory bodies, may emerge as a more effective way to establish answers to fundamental information questions upon which the parties to the issue cannot agree. The need for some sort of

53

SCHUCK, LIMITS OF LAW supra note 15, at 438-440.

54

Sunstein, Informational Regulation, supra note 27; CASS R. SUNSTEIN, FREE M ARKETS AND SOCIAL JUSTICE 327-28 (1997) (discussing connections between regulatory efficiency and informational availablity); Sunstein & Kuran, Availability Cascades, supra note 19 (cautioning against regulatory decisions rooted in false information); Cass R. Sunstein, Reinventing the Regulatory State, 62 U. CHI. L. REV. 1 (1995) (advocating for an increased emphasis on information generation and disclosure in federal regulations); Cass R. Sunstein, Informing America: Risk, Disclosure, and the First Amendment, 20 FLA. ST . U. L. REV. 653 (1993) (discussing the importance of information in effective risk regulation, and stressing the government’s fundamental role in providing it). 55

Coase, Social Costs, supra note 2. Coase himself stressed the importance of needing to tap into unexploited information resources and the distance between the real world and the theoretical Coasean world. See, e.g., Ronald H. Coase, The Institutional Structure of Production, in ESSAYS ON ECONOMICS AND ECONOMISTS 3-14 (1994) [or] Ronald H. Coase, The Institutional Structure of Production, in NOBEL LECTURES IN ECONOMIC SCIENCE II (Torstern Persson ed., 1997). 56

Calabresi, Pointlessness, supra note 2 at 1234 & n.70 (identifying and categorizing barriers to full efficiency); Ellickson, Coaseanism, supra note 2, at 611-612 (pointing out that Coase did not believe that we lived in a “Coasean World”). 57

But note that courts have trouble making such assessments as well. Krier & Schwab, Property Rules, supra note 4.

decisionmaking mechanism or “hierarchy” will be especially salient where the number, complexity, or divergence of values around the gaps is great.58 Because transaction costs are often high and market failures pervasive, a good bit of environmental law focuses on a regulatory model for problem solving rather than on contract or tort approaches to cost internalization. 59 But regulatory strategies for addressing externalities have their own shortcomings. 60 Indeed, in recent years, a great deal of policy analysis has centered on “regulatory reform.”61 Some scholarly commentators have discussed the technical requirements for sound regulatory decisionmaking. 62 But the breadth of information shortcomings in the environmental domain has not been carefully specified, perhaps because the debate over property versus liability rules as the best way to protect entitlements (and especially Calabresi and Melamed’s Cathedral analysis) has cast a long shadow across the academic literature and narrowed thinking on where the critical information gaps lie. In seeking to re-frame the scholarly debate, I highlight an array of foundational data and information issues that are essential to good environmental decisionmaking—but which lie “beneath” the Cathedral discussion.

58

See Williamson’s discussion of “hierarchies” discussed infra at note 84.

59

Steven Shavell, Liability for Harm Versus Regulation of Safety, 13 J. LEGAL STUD. 357 (1984) (identifying determinants of the relative desirability of liability versus regulation). See generally W ILLIAM J. BAUMOL & W ALLACE E. OATES , ENVIRONMENTAL POLICY (1988) (on the economic logic for regulation). 60

Bruce Yandle, Coase, Pigou, and Environmental Rights, in W HO OWNS THE ENVIRONMENT ? (Peter J. Hill & Roger E. Meiners eds., 1998) (explaining why regulatory approaches may do no better than markets in fixing environmental problems); Esty, Revitalizing, supra note 21, at 584-599. 61

DAVIES & M AZUREK, POLLUTION CONTROL, supra note 5 (arguing that environmental regulatory failure stems from fragmented systems of control, complexity of legislative provisions, ineffective administrative remedies , and scarcity of necessary information); A SPEN INSTITUTE, THE A LTERNATIVE PATH: A CLEANER, CHEAPER W AY TO PROTECT AND ENHANCE THE ENVIRONMENT (1996); BREYER, BREAKING, supra note 13; CENTER FOR STRATEGIC AND INT ’L STUDIES , THE ENVIRONMENTAL PROTECTION SYSTEM IN TRANSITION: TOWARD A M ORE DESIRABLE FUTURE (1998); DEBRA S. KNOPMAN, PROGRESSIVE FOUNDATION, SECOND GENERATION—A NEW STRATEGY FOR ENVIRONMENTAL PROTECTION (1996); NATIONAL ENVTL. POLICY INST., REINVENTING THE VEHICLE FOR ENVIRONMENTAL M ANAGEMENT (1995). 62

DAVIES & M AZUREK, POLLUTION CONTROL, supra note 5; LANDY ET AL., A SKING THE W RONG QUESTIONS , supra note 16 (offering explanation for EPA failures during the Reagan years).

1. Property versus Liability Rules Calabresi and Melamed demonstrated that a shift in the basis for protecting entitlements from property rules to liability rules might overcome information gaps related to the proper measure of harm in tort cases. Court-determined damages reduce vulnerability to transaction costs imposed by holdouts and other inefficiencies that arise from strategic bargaining. 63 They specifically showed that with a propertyrule-protected entitlement, the value of the polluting activity blocked by an injunction might be greater than the harms suffered by the pollutees. In a Coasean world, the polluters would compensate the pollutees in these circumstances and still come out ahead. But what if the pollutees are not forthcoming about the value of their injuries? What if they demand unreasonable compensation causing negotiations to break down? Liability rules “collectivize” the damage assessment process. Such an entitlement-protection regime changes the information dynamic, applying court-determined damages in place of negotiation between the parties. The court, in effect, fills the information gap that separates the parties and, in doing so, overcomes any potential negotia ting breakdown. Calabresi further demonstrates that, in a world of less-than-perfect information, the legal regime should be structured to motivate “least-cost-avoiders” to act.64 The Cathedral article and its progeny have sharpened our focus on the ways in which property and liability rules can be utilized, refined, and combined to generate information on the value of harms (and the burdens of abatement by those engaged in harm-causing activities) and to limit strategic bargaining and the inefficiency created by holdouts.65 But as Rose argues, the conclusions drawn from

63

Calabresi & Melamed, Cathedral, supra note 3.

64

GUIDO CALABRESI, THE COST OF ACCIDENTS : A LEGAL AND ECONOMIC A NALYSIS (1970); see also Frank Michaelman, Pollution as a Tort: A Non-Existential Perspective on Calabresi’s Costs, 80 YALE L. J. 647 (1971). 65

See Calabresi & Melamed, Cathedral, supra note 3; Ian Ayres & Eric Talley, Solomonic Bargaining: Dividing A Legal Entitlement To Facilitate Coasean Trade, 104 YALE L. J. 1027 (1995) (spelling out bargaining obstacles); Ian Ayres & Eric Talley, Distinguishing Between Consensual and Nonconsensual Advantages of Liability Rules, 105 YALE L. J. 235 (1995); Ian Ayres & J. Balkin, Legal Entitlements as Auctions: Property Rules, Liability Rules, and Beyond, 106 YALE L. J. 703, 707 (1997) (distinguishing between higher-order liability rules and first-order liability principles); Steven Shavell, Liability for Harm Versus Regulation of Safety, 13 J. LEGAL STUD. (1984) (arguing for liability rules rather than regulation

various contributions to the Cathedral debate are a function of the examples put forward. 66 The structure of the examples chosen thus defines the analysis, assumes away certain issues, and focuses the reader on others.67 Rose notes that the “shadow examples” affect not only the choice of rules but also the content of the entitlements.68 I contend that the shadow of the Cathedral is even more overwhelming, distracting the scholarly community from the broader set of information issues at play in the environmental domain. The fundamental source of this distraction is the use of artificial environmental examples. While Calabresi and Melamed purport to use environmental nuisance to illuminate their argument, they really have in mind the law of accidents. They ignore the foundational issue of whether there was an accident in the first place.69 As I discuss below, in the pollution context, the existence, contours, sources, movement, and impact of the harm in question cannot be taken for granted. 2. Beneath the Cathedral People affected by pollution often do not know what potential environmental injuries they face, where particular harms are coming from, how much these harms affect them, what value to place on the injuries or effects they suffer, nor whether they have a right to be free of the harm. Under these circumstances, a Calabresian bargaining breakdown is the least of their problems. And while some of the

where private parties are in better position to assess risks than government); Louis Kaplow & Steven Shavell, Property Rules Verses Liability Rules: An Economic Analysis, 109 HARV. L. REV. (1996) (explaining why liability rules may be superior to property rules) [hereinafter Kaplow & Shavell, Property Rules]; Carol Rose, Property Rules, Liability Rules, and Inalienability: A Twenty-Five Year Retrospective: The Shadow of the Cathedral, 106 YALE L. J. 2175, 2176 (1997) [hereinafter Rose, Shadow] (suggesting that liability rules require courts to do more than determine the average damages from pollution). 66

Id, at 2178-82 (discussing the “shadow examples”). Rose further notes that contract law is the shadow example in the work of Ayres and Talley and that Kaplow and Shavell again work from the shadow of the law of accidents. 67

Polinsky, for instance, assumes the government has “full information about the externality problem.” A. Mitchell Polinsky, Controlling Externalities and Protecting Entitlements: Property Right, Liability Rule, and Tax-Subsidy Approaches, 8 J. LEGAL STUD. 1 (1979) [hereinafter Polinsky, Controlling]. 68

Rose, Shadow, supra note 65 at 2179. See also Polinsky, Controlling, supra note 67, at 234.

69

Calabresi’s work on accidents in the late 1960s furthermore came against a backdrop of a major policy debate about “nofault” auto insurance, designed to overcome bargaining breakdowns over how to tally damages in the context of car accidents. See ROBERT A. KEETON & JEFFREY O’CONNELL, BASIC PROTECTION FOR THE TRAFFIC VICTIM 181 (1965). Calabresi’s work—and thus his examples—also pre-date the rise of a statutory approach to pollution control.

information failures relate to the incentives created by legal rules, other gaps are more a function of science or values, which require a quite different line of attack to fix. 70 Modern day pollution control problems are, moreover, rarely of the simple “A causes harm to B” type that the property-versus-liability-rules debate explores. Under an even slightly more complex model—with polluters (or harm causers) P, P', and P''; “pollutees”(or victims) V, V', and V"; and pollutants (or emissions) E, E', and E" causing harm X—a much broader array of information failures can be identified. a. Problem Identification In the environmental domain there is often a great deal of uncertainty about whether a problem even exists. In Calabresi’s world, V knows he’s been harmed by P. In the case of an accident, V knows P’s car has hit him causing harm X. Calabresi and other Cathedral commentators also use nuisance examples in which V is assumed to be aware of P’s harm to him. In the broader pollution context, however, V may not see the emissions from P’s factory smokestack. In fact, V may not even see P’s factory, which may be many miles away. Even if V sees the smoke in the air or emissions going into the water, he may not be able to tell if this pollution is harmful. Thus, to establish the existence of a harm, a number of questions must be answered: • Is P emitting E? • Does E cause harm? To complicate this set of “existence” issues, many environmental harms are not identified until many years after they have been generated. Chlorofluorocarbons (CFCs) were considered miracle chemicals in the 1950s with uses that ranged from cleaning semiconductors to refrigeration. Only in the 1970s and 1980s was it recognized that CFCs were depleting the ozone layer.71 The inability to see many problems 70

Some of the problems may also be “beyond science” and related to disputes over values. See Wendy Wagner, The Science Charade in Toxic Risk Regulation, 95 COL. L. REV. 1613 (1995) (separating problems science can address from “trans-science” issues). 71

RICHARD BENEDICK, OZONE DIPLOMACY: NEW DIRECTIONS IN SAFEGUARDING THE PLANET 9-22 (1998) (summarizing the development of the scientific knowledge regarding CFC’s and depletion of the ozone layer).

arises from the fact that modern-day pollutants are often, quite literally, invisible. Other harms are spatially or temporally diffused. Problems may not emerge until an ecosystem’s assimilative capacity is overwhelmed or other critical thresholds are crossed. The invisible buildup of CO2 and other greenhouse gases in the atmosphere potentially causing climate change over a period of decades represents the paradigmatic example of the hidden harm that spills across space and over time. To the harm identification taxonomy, we must therefore add: • Can V (or V' or V") perceive E? b. Causal Specification Even if the existence of a potential harm is clear, important questions may remain about the impacts of the harm-causing emissions in question—and thus the potential for damage to people or the natural environment. In the ecological sciences, these issues are studied as matters of the “fate and transport” of pollutants. 72 In the law, these questions are analyzed as a matter of causal connections. In any case, we need information to establish a link from particular emissions to specific receptors or pollutees: • Does P’s E affect V? In reality, the information issues and uncertainties surrounding causal linkages are even more complex. Notably, V may be affected by emissions from sources other than P. Are the particulates P is breathing from the power plant next door? The factory across town? The cars miles upwind? Thus, a comprehensive data base must identify additional sources of E such as P' and P''. Moreover, P’s emissions (as well as P'’s and P''’s) may well affect not only V but V' and V". We therefore have several additional issues to address: • Do the emissions from P' and P" affect V? • Does P’s E affect V' and V"?

72

See, e.g., R. W ATTS, HAZARDOUS W ASTES : SOURCES , PATHWAYS , RECEPTORS (1998); JERALD L. SCHNOOR , ENVIRONMENTAL M ODELING, FATE AND TRANSPORT OF POLLUTANTS IN W ATER, A IR AND SOIL (1996).

c. Impact Evaluation: Epidemiological and Ecological Effects Even when we can trace emissions of E from P to V, uncertainties about how the pollution affects the pollutee may persist. In the policy domain these issues loom large. Does smog hurt people?73 Do radionuclides in drinking water lead to cancer?74 Obviously, mere physical contact does not necessarily create a need for redress or intervention. Only actual damage or the risk of injury requires action. Information about the epidemiological and ecological impacts of E must thus be developed, analyzed, and disseminated. To our catalogue of information needs, we must add: • Does E harm V? • If so, how? • What dose of E affects V? The issue of epidemiological and ecological effects may be further complicated by a number of factors. First, the impacts on V may not occur immediately. As noted above, threshold effects may only emerge after a particular level of exposure or intensity of impact has been realized. A single car in Los Angeles is not a problem; the wind dissipates the exhaust. But five million cars overwhelms the assimilative capacity of the airshed. In other circumstances, harms accrue over time and only become visible after a certain degree of accumulation. The ozone layer only began to breakdown after the buildup of CFCs in the stratosphere exceeded a certain level. In addition, the impact on V may be indirect. For instance, pesticide exposures may come through residue on food. Second, V’s injuries might be a result of exposure to other pollutants beyond E. Is respiratory distress a function of smog? Or rather particulates? Or air-borne toxics? This multi-dimensionality means we must ask: • Do V’s symptoms come from exposure to E' or E" as well as to E? 73

EPA, SMOG—W HO DOES IT HURT ? (Jul. 1999), available at http://www.epa.gov/airnow/health/ (discussing ongoing uncertainties with regard to the ecological and epidemiological effects of smog). 74

National Groundwater Association, Radionuclides: What You Need to Know, at http://www.ngwa.org/pdf/Radiocuclides.pdf (highlighting the risks and uncertainties posed by these naturally occurring radioactive contaminants).

Third, a further set of information needs emerge at the intersection of questions about effects (both epidemiological and ecological) and causal uncertainties. Specifically, particular subgroups of the population (e.g., children, older people, those with compromised immune systems) may all need to be analyzed separately. Moreover, some injuries may be a function of the interaction of one pollutant with another or with other risk factors. Exposure to radionuclides is, for example, much riskier for smokers than non-smokers. The potential for interactive effects means we must explore: • Do V’s symptoms come from the interaction of E with E' and E"? • Are there other factors that aggravate or mitigate E’s impact on V?75 d. Harm Valuation Even if we are able to ascertain the precise physical effects of an identified source of pollution on a pollutee, we face the additional question of trying to calculate the dollar value of the injuries inflicted. It is at this point that the Cathedral literature picks up. Getting the valuation of harms right is critical— and often subject to real uncertainty. 76 The valuation calculus raises a number of questions: • Who should undertake the valuation? • Is technical expertise required? • Whose values and valuations should be used? Those of the pollutee? Of society? • Should we fully compensate victims who are unusually susceptible to harm (e.g., those with weak respiratory or immune systems)? • Should we fully compensate those whose own behavior (e.g., smoking) creates the risk of injury or worsens the damage?

75

Age, weight, sex and other factors such as demographic groups with unusual dietary patterns (e.g., Native Americans who eat a great deal more fish than the rest of the population) may all be relevant variables to examine. JOHN W ARGO, OUR CHILDREN’S TOXIC LEGACY (1998) (spelling out how and why some subgroups of the population may be more sensitive to certain harms). 76

Polinsky observes that difficulties in calculating damages could be more important than strategic behavior as a source of breakdown in the Calabresi-Melamed model. A. Mitchell Polinsky, Resolving Nuisance Disputes: The Simple Economics of Injunctive and Damage Remedies, 32 STAN. L. REV. 1075, 1111 (1980). Krier and Schwab similarly note that judicial efforts to place a value on harms under liability rules often entail substantial “assessment costs.” Krier & Schwab, Property Rules, supra note 4.

• How do we discipline exaggeration or other strategic behavior? The multi-person, multi-element model again reveals additional complexities that add to our information needs and burdens. Specifically, if V is affected by E from P' and P'' as well as P, how do we determine what share or which units of harm P is responsible for? This issue is particularly important in circumstances where the marginal cost of the harm is not steady but rising (or falling). 77 The damage assessment questions highlighted by the Cathedral debate are thus significant but cannot be addressed without reference to a broader set of underlying information needs that arise in any environmental decisionmaking context. e. Rights Delineation Coasean analysis focuses on what happens after an entitlement has been allocated. The Cathedral debate similarly presumes that there has been an allocation of entitlements prior to the exploration of whether property or liability rules should be deployed. But in the environmental realm, the delineation of property rights is often a contentious issue,78 requiring us to ask. • Have the relevant property rights been established? • Are the parties aware of who holds the rights? • Are the boundaries clear? Where the rights have been allocated, the inquiry turns to a second set of questions: • Has there been an encroachment on these rights? • Can the rights holder tell? • Are the rights easily vindicated?

77

Rose, Shadow, supra note 65, at 2176.

78

One dimension of this issue centers on the fairness of the initial allocation of rights. This Article elides this distributional justice issue. As others have noted, without diminishing the seriousness of equity concerns, other more direct policy approaches (i.e., progressive taxation) are likely to better achieve any desired distributional outcome. See, e.g., Kaplow & Shavell, Property Rules, supra note 65, at 744.

While Calabresi implicitly (and other contributors to the Cathedral literature more explicitly 79 ) assumes that the ownership, scope, and boundaries of the relevant property rights are clear, in the real world of environmental decisionmaking, they often are not. Considerable effort often goes into ascertaining who holds the relevant rights and whether an externality exists that interferes with these rights, highlighting another potential information gap and source of transaction costs.80 What constitutes an externality is also a function of values and community norms, which makes the scope of rights fluid and subject to further uncertainties and potential information gaps.81 f. Policy Intervention Once an environmental harm has been identified and deemed worthy of attention, two other sets of questions emerge, centered on: (1) abatement options and costs and (2) who should take action. In many circumstances, an array of pollution control alternatives need to be considered: restructured processes, redesigned products, pollution prevention options, “end-of-pipe” treatment, and so on. Protecting the pollutees or moving them should also be considered. Putting the incentives in the right place to generate an optimal degree of creativity about the policy options and how to minimize the cost of intervention has proven to be critical. 82 Determining whether P or V is better positioned to reduce the harm provides a useful starting point but not an endpoint for this effort. Given the scientific and technical complexity of many modern-day environmental problems, neither P nor V may really be in a positio n even to understand the options for abatement of E. Thus, we need to widen our vision about who might be motivated to explore the opportunities to reduce the harms in question—and how to optimize their 79

Polinsky, Controlling, supra note 67, at 13.

80

Paul R. Portney, EPA and the Evolution of Federal Regulation, in PUBLIC POLICIES FOR ENVIRONMENTAL PROTECTION 7, 11-12 (Paul R. Portney ed., 1990) (arguing that nonregulatory approaches are insufficient due to difficulties in defining rights, prohibitive transaction costs, and market imperfections); Polinsky, Controlling, supra note 67; Carolyn Woj, Property Rights Disputes: Current Fallacies and a New Approach, 14 J. LEGAL STUD. 411 (1985). 81

See Robert E. Ellickson, Suburban Growth Controls: An Economic and Legal Analysis, 86 YALE L. J. 385, 429, 475-89 (1977); Robert E. Ellickson, Alternatives to Zoning: Covenants, Nuisance Rules, and Fines as Land Use Controls, 40 U. CHI. L. REV. 681, 728-731 (1973) (noting that evaluative terms such as beneficial and harmful are often made with reference to community determined standards of conduct).

motivation. We also need to generate information on the costs of the alternative actions. Finally, we need to determine who is best positioned to act on the opportunities identified and to decide which option to pursue. These added dimensions of uncertainty create new information burdens: • How do we best reduce the effects of E? What are our policy choices? • What can be done to motivate innovation in pollution control? • Who is best positioned to act to reduce E? How do we minimize system costs? • What institutional design is best suited to the problem to be addressed? g. Implementation Once an optimal abatement strategy has been identified, somebody must act to implement the chosen intervention. In general, these “action” costs will be borne by a private party. But in the regulatory model, a government agency may share action responsibilities. Thus, an additional set of issues arise around how to minimize the costs of administering the intervention strategy: • Is the intervention selected implemented effectively and efficiently? • How can administrative costs be minimized? h. Monitoring and Enforcement Even after we have assigned responsibility for reducing a pollution harm, there remain important questions about whether this party carries out the assigned responsibilities. Courts must ascertain whether their orders are obeyed—and intervene where parties do not comply. Likewise, a great deal of time and energy is spent within environmental agencies tracking the compliance of polluters. These monitoring and enforcement issues mean that we must add to our list of potential areas for information failure questions such as: • Did those assigned to reduce harm X do what they were supposed to do? • How do we best track compliance with our program of E control?

82

Michael E. Porter, America’s Green Strategy, 4 SCI. A M. 264 (1991) (arguing that performance standards rather than technology mandates are the key to innovation).

i. Updating and Refinement Given the dynamic nature of environmental problems and the fast pace at which knowledge evolves in the ecological sciences, each element of the foregoing analysis must be constantly reexamined and refined. Changes in circumstances—increased population density, greater scarcity of resources, shifts in the underlying scientific knowledge base, or technological breakthroughs— will often change the answers to questions somewhere along the information track that has just been laid out. To our taxonomy of potential information failures the following set of issues must therefore be added: • Has anyone come up with a better way to deal with harm X? • Does this knowledge breakthrough change our assumptions about: - the existence of X - fate and transport of E causing X - epidemiological or ecological effects of E - valuation to be placed on X - who holds the relevant property rights - intervention options for dealing with X - how best to monitor compliance with E abatement efforts Table 1 summarizes the pollution control “information needs” taxonomy:

TABLE 1: Taxonomy of Environmental Problem Solving Information Needs

Problem Identification:

Is P emitting E? Does E cause harm? Can V (or V' or V") perceive E?

Causal Specification:

Does P’s E affect V? Does the emissions from P' and P affect V? Does P’s E affect V' and V"?

Impact Evaluation:

Does E harm V? If so, how? What dose of E affects V? Do V’s symptoms come from exposure to E' or E" as well as E? Do V’s symptoms come from the interaction of E with E' and E"? Are there other factors that aggravate or mitigate E’s impact on V?

Harm Valuations:

Who should undertake the valuation? Is technical expertise required? Whose values and valuations should be used? The pollutee’s? Society’s? Should we fully compensate victims who are unusually susceptible to harm (e.g., those with weak respiratory or immune systems)? Should we fully compensate those whose own behavior (e.g. smoking creates the risk of injury or worsens the damage? How do we discipline exaggeration or other strategic behavior? How much value should we place on the effects of P’s E on V?

Rights Delineation:

Have the relevant property rights been established? Are the parties aware of who holds the rights? Are the boundaries clear? Has there been an encroachment on these rights? Can the rights holder tell? Are the rights easily vindicated and enforced?

Policy Intervention:

How do we best reduce the effects of E? What are our policy choices? What can be done to motivate innovation in pollution control? Who is best positioned to act to reduce E? How do we minimize system costs? What institutional design is best suited to the problem to be addressed?

Implementation:

Is the intervention selected implemented effectively and efficiently? How can administrative costs be minimized?

Monitoring and Enforcement:

Did those assigned to reduce harm X do what they were supposed to do? How do we best track compliance with our program of E control?

Updating and Refinement:

Has anyone come up with a better way to deal with the harm? Does this knowledge breakthrough change our assumptions about: • the existence of X? • fate and transport of E causing X? • epidemiological or ecolo gical effects of E? • evaluation to be placed on X? • who holds the relevant property rights? • intervention options for dealing with X? • how best to monitor compliance with E abatement efforts?

Where: Harm Polluters Pollutees pollutants

= = = =

X P, P', P'' V, V', and V" E, E', and E'

As demonstrated here, the spectrum of potential information failures threatening to disrupt efficient cost internalization is much broader than the Cathedral debate might lead readers to belie ve. Indeed, the range of issues may seem daunting. The sweep of concerns certainly argues for much more scholarly attention (and policy focus) on categorizing and analyzing how to address various information gaps.

III. Gapfilling and Institutional Design While Calabresi and Melamed disclaim any focus on “what institutions and what procedures are most suitable for making what decisions,” 83 other scholars highlight the issue of institutional design as the focus of efforts to overcome information failures. Working within the framework of corporate organization and expressly putting aside the Coasean “fiction of frictionless” transactions, Williamson observes that every contract is to some degree incomplete. “Gapfilling” to fill these information holes becomes essential to complete the transaction. 84 How voids are filled and whether the institutions used to generate the needed information minimize transaction costs thus emerges as a central determinant of economic efficiency. In the context of a firm, Williamson identifies two fundamental gapfilling options: markets or hierarchies.85 Because Williamson’s analysis centers on corporate organization, his transaction-cost-based categorization of institutional structures requires some translation to make sense in the environmental realm. The relevant “transaction” is not that of a firm but the effort to conclude Coasean bargains over environmental property rights or, more broadly, to internalize environmental externalities. A slightly different set of institutio nal options thus emerges: (a) markets and (b) non-market structures including tort law and courts as well as various types of governmental regulation. Although not strictly a “structure” for 83

Calabresi & Melamed, Cathedral, supra note 3, at n.2.

84

OLIVER W ILLIAMSON, M ARKETS AND HIERARCHIES : A NALYSIS AND A NTITRUST IMPLICATIONS 139, 253-54 (1975) [hereinafter W ILLIAMSON, M ARKETS AND HIERARCHIES ]; Oliver Williamson, Transaction Cost Economics: The Governance of Contractual Relations, 22 J. L. & ECON. 3 (1979) [hereinafter Williamson, Transaction Cost Economics] (noting how transaction costs affect outcomes); Oliver E. Williamson, Transaction Cost Economics, in 1 HANDBOOK OF INDUSTRIAL ORG. 135 (Richard Schmalensee & Robert E. Willy eds., 1982). 85

W ILLIAMSON, M ARKETS AND HIERARCHIES , supra note 84 at 139-254.

internalizing harms, we might also examine the pressure of public mores and the social context of environmental concern they create. Further separation of regulation into “command and control” mandates, market-based mechanisms, and “command and covenant” 86 procedures might also be considered. The set of core institutional options for addressing environmental problems therefore includes: (1) contractual exchange; (2) tort liability; (3) regulation — (a) command and control regulation; (b) market-based mechanisms; (c) command and covenant procedures; and (4) social context. Williamson’s transaction cost emphasis highlights the fact that we cannot achieve optimal policy results simply by comparing the costs of taking action (pollution control) with those of inaction (pollution harms left unabated). Economic efficiency requires a focus on another fundamental question: which structure best fills information gaps and thus minimizes transaction costs?

86

See E. Donald Elliott, Toward Ecological Law and Policy, in THINKING ECOLOGICALLY, supra note 44 (explaining how a “command and covenant” approach to environmental protection might work).

Table 2: Least Social Cost Environmental Protection — seeks to minimize G (A + I + T), where: A= I = T =

action costs (i.e., pollution control costs or user costs) inaction costs (i.e., failure costs, unabated harms borne by pollution victims) transaction costs, which encompass: –harm identification –causal specification –impact evaluation (epidemiological and ecological effects) –harm valuation –rights delineation –intervention options and benefit/cost assessments –implementation including: •negotiation costs (contract) •court costs (tort) •policy determinations (regulation) –monitoring and enforcement –updating and reinforcement

It turns out that, just as the nature of the transaction determines which form of corporate organization is most efficient, the nature of the environmental problem determines which cost internalization structure is most efficient. Under a least-social-cost metric, high gapfilling costs (T) can swamp the benefits of pollution control (reduced I). Indeed, in a significant number of circumstances today, the costs of intervention to address environmental harms (A+T) are judged to be too high relative to expected gains (reduced I) to justify taking action. 87 More generally, as information costs fall, the gapfilling problem will be transformed — leading us toward new optimal regulatory strategies. A. Obstacles to Gapfilling Before turning to the question of which institutional structures most efficiently fill the gaps in environmental decisionmaking, it is important to understand how the gapfilling process might break down. In this regard, the following taxonomy of potential gapfilling failures is useful to keep in mind.

87

The broader implications of adopting the efficiency emphasis of a least social cost approach to environmental protection are discussed in sub-parts C and D below.

1. Technical Deficiencies Some decisionmaking gaps involve direct information failures where essential technical data or analysis on which to base a decision is not available. In some cases the requisite information does not exist. Science may not have advanced sufficiently to answer a critical question. For example, how much acid rain causes harm to fish? In such cases, as Rose points out, the challenge is to create incentives to generate the missing information. 88 The government might invest in scientific research to fill the gaps that are present.89 Alternatively, the law might be structured in ways that create an incentive for the least-cost-informationgenerators — often producers — to do the requisite analysis. But producers are frequently the polluters as well and, under current legal structures, they frequently have little incentive to generate information on the harms they are causing. Incentives for ignorance are thus sometimes to blame for lack of data on which to base pollution control decisions. 90 In some other cases, the information needed to close a gap exists but is not being brought to bear on the decision process. For example, factory owners or managers often know where the greatest environmental risks are at their facilities, but they see no reason to bring this information forward. 91 Once again, we need to ask: what incentives are required to get the information from those who have it to the

88

See Carol Rose, Scientific Innovation and Environmental Protection: Some Ethical Considerations, 32 ENVTL. LAW 755, 759-64 (2002). 89

With regard to acid rain, for example, the government spent hundreds of millions of dollars attempting to establish the scope of the problem. See the website of the National Acid Precipitation Assessment Program, at www.oar.noaa.gov/organization/napap.html. 90

The Toxic Substances Control Act (TSCA) § 6(a), 15 U.S.C. §2605(a), for example, puts the burden on the EPA to identify a substance’s health and environmental risks before seeking to regulate it thereby giving the producer a free pass and an incentive to discourage EPA’s information creation; see also Holly E. Pettit, Shifting the Experiment to the Lab: Does EPA Have a Mandatory Duty to Require Chemical Testing for Endocrine Disruption Effects Under the Toxic Substances Control Act, 30 ENVTL. L. 413 (2000) (discussing TSCA’s burden of proof). 91

E. Donald Elliott & Gail Charney, Toward Bigger Bubbles: Why Interpollutant and Interrisk Trading are Good Ideas and How We Get There from Here, 13 F. A PPLIED RES . & PUB . POL’Y. 48 (Winter 1998) [hereinafter Elliott & Charney, Bigger Bubbles].

relevant decisionmakers? Such a structure of incentives might again involve burden shifting as is done under California’s Toxic Regulation Program (Proposition 65).92 2. Cognitive Shortcomings Another set of failures emerges from decisionmakers’ inability to make proper sense of available data. The human brain is a limited information processor and often cannot manage to take appropriate note of the different stimuli that are crossing the perceptual threshold. 93 To contend with vast arrays of information and process it in an efficient manner, humans rely on mental shortcuts or “heuristics” to facilitate rapid information processing. And they employ organizing principles or “schema” to assist in the identification and classification of relevant information. Psychologists have also begun to explain how emotional (affective) and societal (collective) factors interact with data processing in the human brain to shape risk assessment and decisionmaking processes. Although designed for efficiency, these sorts of information-processing strategies often yield systematic and predictable errors which, when magnified on a societal level, can severely distort the regulatory process. “Law and psychology” scholars have developed a taxonomy of these heuristics — such as availability, affect, probability neglect, threshold and certainty effects and so on — in an attempt to understand common deficiencies in human cognition. 94 Recognizing the social cost of

92

Proposition 65 simply requires businesses to provide warnings whenever their products or activities may cause exposure to listed carcinogens and reproductive toxins above de minimus levels. There fore, although the law does not prevent companies from releasing hazardous products, it does contain serious penalties for failure to adequately warn. This threat of liability creates powerful incentives for companies to generate the information necessary to determine safe exposure levels, and to furnish the government with this information in the hopes that it will establish thresholds of “no significant risk.” This “burden-shifting” thereby effectively eliminates the obstructionism that oftentimes characterizes the information-gathering effort. See, e.g., Bradley C. Karkkainen, Information as Environmental Regulation: TRI and Performance Benchmarking, Precursor to a New Paradigm? , 89 GEO. L. J. 257 (2001) (praising the information generating power of Prop 65 and other “burden-shifting” statutes); Michael W. Graf, Regulating Pesticide Pollution in California Under the 1986 Safe Drinking Water and Toxic Exposure Act (Proposition 65), 28 ECOLOGY L. Q. 663 (2001) (hailing Prop 65’s informational and pollution-reducing potential, and advocating that its discharge and disclosure requirements be extended to pesticides and other under-regulated chemicals). 93

Rachlinski & Farina, Cognitive Psychology, supra note 19, at 555 (2002) (describing the brain’s limitations as an information processor and the basic mental tools humans have adopted so as to compensate and conserve scarce cognitive resources). 94

See, e.g ., SLOVIC , PERCEPTION OF RISK, supra note 19 (suggesting the existence of “availability” and “affect” heuristics, leading to the systematic overestimation of high profile or intensely emotional risks); Cass R. Sunstein, Book

these errors, a growing literature has emerged to explain human cognitive failures, catalogue the various biases that arise, and propose strategies for their remediation. 95 Risk calculations do not occur in a “social vacuum.” 96 They are heavily influenced by societal factors and pressures. Slovic, Jolls, Sunstein and others, have identified numerous social mechanisms that serve to amplify or attenuate societal perception of risks. 97 For example , an individual’s risk perception will ofte n be significantly influenced by the “availability” of the issue in public discourse and how others perceive the threat. As people generally have little first-hand knowledge about the likelihood of various risks and rarely independently research the dangers, they rely on signals derived from discussions with peers, articulation by public figures, or repetition in the media and public domain. “Availability entrepreneurs,” activists who understand the social processes that induce collective belief formation, will often try to capitalize on these social mechanisms in an effort to trigger availability “cascades” likely to advance their own agendas.98

In

addition, in a process known as “group polarization,” like-minded individuals will often gravitate towards each other, reducing the range of opinions and arguments within a given discussion group. This dynamic may set in motion processes of collective reinforcement that yield polarized viewpoints, resulting in unjustified alarmism or indifference.99

Review, The Perception of Risk , 115 HARV. L. REV . 1119, 1138 (2002); (summarizing Slovic’s central propositions); and Cass R. Sunstein, Probability Neglect: Emotions, Worst Cases, and Law, 112 YALE L.J. 61 (2002) [hereinafter Sunstein, Probability Neglect] (describing the tendency to “neglect” probabilities when intense emotions are engaged, arbitrary safe/unsafe thresholds are crossed, or out of a psychic need to create the semblance of “certainty”). 95

Id. See generally Amos Tversky & Daniel Kahneman, Judgment Under Uncertainty: Heuristics and Biases (Daniel Kahneman et al. eds., 1982); Sunstein & Kuran, Availability Cascades, supra note 19; Roger G. Noll & James E. Krier, Some Implications of Cognitive Psychology for Risk Regulation, 19 J. LEGAL STUD. 747, 769-71 (1990); Jolls et al., Behavioral Approach, supra note 10; HEURISTICS AND BIASES : INTUITIVE JUDGMENT (Thomas Gilovich et al. eds.) (2002). 96

SLOVIC , PERCEPTION OF RISK, supra note 19.

97

Id.; Christine Jolls, Book Review ‘Social Behavioral’ Economics, 5 GREEN BAG 2d 321 (2002); Jolls et al., Behavioral Approach, supra note 10; Sunstein & Kuran, Availability Cascades, supra note 19 (developing the “availability cascade” concept). 98

Sunstein & Kuran, Availability Cascades, supra note 19, at 683.

99

Cass R. Sunstein, Deliberative Trouble, Why Groups Go to Extremes, 110 YALE L.J. 71 (2000).

These methods of societal risk analysis and collective belief formation interact with basic cognitive processes and failures, often resulting in serious risk distortions and chronic errors of overor under-regulation. Politicians, themselves, are in fact, uniquely susceptible to certain forms of cognitive biases and social influences, as are the experts and officials who draft many of our regulations. 100 Cognitive failures thus represent a serious stumbling block on the path toward appropriately filling environmental information gaps. 3. Structural Mismatches Sound environmental decisionmaking requires that all the costs and benefits of a potential action (or inaction) be tabulated and factored into the decision calculus. Where the decisionmaker’s jurisdiction or purview does not encompass all cost-bearers or beneficiaries, suboptimal choices will be made.101 Polluting factories may be under-regulated as air pollution harms which drift downwind beyond state borders tend to be ignored. 102 When transboundary effects are disregarded, the decision process suffers a “structural” failure — and externalities or internalities103 emerge, distorting regulatory outcomes.104 Structural failures are not per se information failures. But where the jurisdiction of the regulator does not

100

See Rachlinski & Farina, Cognitive Psychology, supra note 19, Part II, for an exposition of the particular fallibilities of experts and elected officials. 101

E. Donald Elliott, Bruce A. Ackerman & John C. Millan, Toward a Theory of Statutory Evolution: The Federalization of Environmental Law, 1 J. L. ECON. & ORG. 313 (1985) (demonstrating the need for alignment between cost bearers and beneficiaries of regulation). 102

See, e.g., the efforts of the Ozone Transport Assessment Group (OTAG) to confront the problem of Midwestern power plants polluting Eastern States. See also Patricia R. McCubbin, Looking to Upwind States to Reduce Interstate Ozone Pollution, 31 ENVTL. L. REP . 11045 (2001); Christina C. Caplan, The Failure of Current Legal and Regulatory Mechanisms to Control Interstate Ozone Transport, 28 ECOLOGY L.Q. 169 (2002); and the website of OTAG, at http://www.epa.gov/ttn/naaqs/ozone/rto.html. 103

See Mancur Olson, Jr., Strategic Theory and Its Applications — The Principle of “Fiscal Equivalence”: The Division of Responsibilities Among Different Levels of Government, 59 A M. ECON. REV. 479, 482 (1969) [hereinafter Olson, Strategic Theory] (observing that “internalities” arise when a public good reaches only a subset of the population in a jurisdiction). 104

Henry N. Butler & Jonathan R. Macey, Externalities and the Matching Principle: The Case for Reallocating Environmental Regulatory Authority, 14 YALE L. & POL’Y. REV. 23 (1996) [hereinafter Butler & Macey, Matching Principle]. See also Esty, Revitalizing, supra note 21, at 587-592.

match the scope of the harm in question, the costs or benefits beyond the sphere of vision of the authorities can be seen as an information gap. 4. Public Choice Distortions A burgeoning literature spells out the range of ways that public choice failures affect regulatory decisionmaking. 105 In some cases, special interests distort the decision processes by investing in campaign contributions, lobbying, or illegal gratuities designed to change the position of decisionmakers. In other cases, the asymmetry of positions and incentives for political involvement among interested parties will distort outcomes because smaller homogeneous groups with visible and significant stakes in the decision tend to be easier to organize than larger, more heterogeneous groups.106 Thus, in many circumstances, polluters with relatively large stakes in environmental policy choices will be wellpositioned to get what they want from the political process while the public, with low per capita stakes, will not be.107 In other cases, self-serving bureaucrats may make choices that maximize their own longterm interest rather than those of society at large. This process could result simply in inefficiency or overregulation as officials try to wield influence, or under-regulation as they neglect their duties in favor of other goals such as leisure time.108 Once again, information gaps may not be the underlying source of

105

See generally M ASHAW , GREED, supra note 17; DANIEL A. FARBER & PHILIP P. FRICKEY, LAW & PUBLIC CHOICE: A CRITICAL INTRODUCTION (1991); Matthew D. McCubbins, Roger G. Noll, & Barry R. Weingast, Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, 75 VA. L. REV. 431 (1989) (illuminating linkages b etween administrative agency structure and political control); Richard B. Stewart, The Reformation of American Administrative Law, 88 HARV. L. REV . 1667, 1684-87, 1713-15 (1975) (detailing widespread industry manipulation of the regulatory process); William Eskridge, Jr., Politics Without Romance: Implications of Public Choice Theory for Statutory Interpretation, 74 VA. L. REV. (1988) (chronicling pervasive subversion of the administrative process for private gain). 106

GORDON TULLOCK, THE CALCULUS OF CONSENT (1962) (spelling out this logic); Roger G. Noll, Economic Perspectives on the Politics of Regulation, in HANDBOOK OF INDUSTRIAL ORGANIZATION 1265 (Richard Schmalensee & Robert D. Willig eds., 1989) (explaining why concentrated interests have an advantage over diffuse ones). 107

In some cases, the benefits of regulation will be concentrated and the costs of intervention diffuse, leading to overrather than under-regulation. In fact, NIMBYism (“not in my backyard”) can be seen as an example of this dynamic. James T. Hamilton, Missing the Mark(et) in Siting Hazardous Waste Facilities, 1 DUKE ENVTL. L. & POL’Y. F. 11 (1991) (arguing that neighborhoods’ abilities to organize is a significant factor in the siting of hazardous waste facilities); Daniel Mazmanian & David Morell, The “NIMBY” Syndrome: Facility Siting and the Failure of Democratic Discourse, in ENVIRONMENTAL POLICY IN THE 1990S : TOWARD A NEW A GENDA 125, 126-27 (Norman J. Vig & Michael E. Kraft eds., 1990) (describing causes of NIMBY syndrome) [hereinafter ENVIRONMENTAL POLICY IN THE 1990S ]. 108

JAMES Q. W ILSON, BUREAUCRACY: W HAT GOVERNMENT A GENCIES DO AND W HY THEY DO IT (1989) (spelling out the pathologies of bureaucracies); Roger G. Noll, Government Regulatory Behavior: A Multidisciplinary Survey and

public choice failures and related decisionmaking shortcomings. But more readily available information — exposing policymaking distortions and facilitating performance comparisons across decisionmaking entities — may help to ameliorate the problem. 109 5. Administrative Inefficiency A variation on the theme of the self-serving bureaucrat is the concern that governmental processes are not structured to achieve efficiency. 110 As a result, information may be processed, but much more slowly and ineffectively than would otherwise be possible. Once again the issue becomes a question of the incentives placed in front of decisionmakers to perform quickly , fairly, and effectively. As noted earlier, regulatory competition has emerged as a critical strategy for promoting governmental efficiency. 111 As I discuss in Part V below, comparative data can be used to generate competitiveness pressures that will highlight the fact that a particular agency is operating inefficiently and that other jurisdictions are getting more bang for their regulatory buck. B. Collectivization v. Decentralization Whether to centralize or decentralize the gapfilling process represents a fundamental policy choice. Because every environmental issue is to some degree unique, there is a logic to case-by-case, individualized analysis. 112 Decisionmaking on such a granular basis, however, entails substantial information processing costs. Thus, just as the shift from property rules to liability rules achieves

Synthesis, in REGULATORY POLICY AND THE SOCIAL SCIENCES 9 (Roger G. Noll, ed. 1985) (examining common bureaucratic sources of regulatory failure). 109

See Part V below for a more complete discussion of the value of comparative data as an antidote to p ublic choice distortions. 110

Daniel C. Esty, Why Measurement Matters, in ENVIRONMENTAL PERFORMANCE M EASUREMENT : THE GLOBAL REPORT 2001-2002 (Daniel C. Esty & Peter Cornelius, eds., 2002) [hereinafter ESTY & CORNELIUS , ENVIRONMENTAL PERFORMANCE M EASUREMENT ]. 111

112

See articles cited at footnote 21, supra .

But the administrative cost of having a governmental structure that matches all these levels is untenable. Esty, Environmental Governance, supra note 9, at 1554-56 (developing a theory of “optimal environmental areas” for regulation building on the “optimal currency area” literature).

efficiency gains by replacing individual damage claims with court-determined awards, regulation “collectivizes” information-intensive aspects of the environmental decisionmaking process. The shift from individual scale analysis to community (or state or national) scale analysis may alleviate the heavy information burden of particularized fact-finding and decisionmaking in several ways. First, negotiation by polluters and pollutees (who may not be technically capable) is displaced by regulatory agency officials who are presumed to be adept at information gathering and analysis. Their expertise may translate into faster analysis and lower costs. Second, shifting the problem context from case-specific to generic circumstances changes the information needs equation. Under the regulatory model, source identification, rights delineation, impact assessment, harm valuation, and intervention option assessment are all undertaken at an aggregate rather than individual scale, reducing the high cost of a particularized inquir ies into each of these issues. This shift in the scope of analysis substitutes average emissions exposures and impacts and generalized costbenefit calculations for individualized preferences, values, and judgments. Collective analysis and decisionmaking thus generates significant economies of scale. Third, the shift toward a regulatory model economizes on information costs through repetition, learning, and comparative analysis. When the decisionmakers (i.e., environment agency officials) address similar problems again and again, they learn from experience and build a base of knowledge, which translates into information gapfilling efficiencies (e.g., faster harm recognition, greater speed of analysis, and so on). They also can use their over-arching vantage point to spot trends, identify “best practices” and disseminate preferred policy strategies. Finally, a regulatory approach often relies on generalized patterns of emissions or harm—and allows officials to assert a causal nexus between pollution and injuries where the particularized facts might have been too limited to support a legal claim under the law of tort.113 This statutorily authorized shortcut in causal specification may substantially reduce overall information requirements.

113

For instance, under the federal Superfund statute, the joint and several liability provision can mean that companies are held responsible for cleanup costs without showing that their activities actually caused the damage in question. CERCLA

The regulatory model also shifts the burden of information production. Contract and tort regimes place the primary burden on individuals. Pollution victims must generate the requisite decision data to open negotiation for compensation or launch a legal case — and the polluters must respond. A regulatory approach to environmental protection shifts much of the pollutees’ (and sometimes the polluter’s) burden to the government. This shift in who bears the brunt of the pollution control transaction costs clearly affects the efficiency of the institutional approach. But it also has equity implications. To the extent that “innocent” victims of pollution harms are relieved of the burden of “making a case,” a regulatory approach may be seen as more fair. However, as I discuss in greater detail below, the transaction cost relief may have other distributional consequences that are less favorable. Notably, the tendency toward using gross averages almost certainly leaves some victims facing above average pollution harms and/or getting under-compensated. Table 3 distinguishes among institutional approaches to environmental protection on the basis of who bears the information burden. Note that while a command and control regulatory regime reduces the analytic burden on each individual, the cumulative information needs remain high. Governments may be better positioned than pollutees to generate the required data and analysis, but they will not always be the least-cost information gapfiller. As a result, a “hybrid” strategy that centralizes some functions and decentralizes others will often minimize overall information costs. And in any case, regulatory approaches need to be designed with much more thought given to what information is needed for a sound decision and who was best positioned to provide this information.

§ 107, 42 U.S.C. § 9607 (West 2003). Similar softening of the causal link between emissions and harm can be found in the Clean Air and Clean Water Acts. John C. Nagle, CERCLA, Causation and Responsibility, 78 M INN. L. REV. 1493 (1994), (discussing the “uneasy coexistence” between causation and many environmental statutes).

TABLE 3: WHO B EARS RESPONSIBILITY FOR INFORMATION GENERATION UNDER VARIOUS INSTITUTIONAL APPROACHES Issue:

CONTRACT

TORT

COMMAND AND CONTROL REGULATION

MARKETBASED REGULATION

“COMMAND AND COVENANT”

SOCIAL CONTEXT (“Voluntary Programs”)

Problem Identification

victim

victim

government

government

government or harm-causer

harm-causer

Causal Specification

victim

victim

government

government

government or harm-causer

harm-causer

Impact Evaluation:

victim/harm-causer

victim

government

government and/or harm-causer

government or harm-causer

harm-causer

Harm Valuation

victim/harm-causer

court

government

Government and/or harm-causer

government or harm-causer

harm-causer and/or government

Rights Delineation

victim/harm-causer

court

government

government

negotiated

harm-causer and/or government

Intervention Options and Costs

victim and/or harm-causer

victim/ harm-causer

government

harm-causer

harm-causer

harm-causer

Implementation

victim and/or harm-causer

victim/ harm-causer

government/ harmcauser

harm-causer

harm-causer

harm-causer

Monitoring and Enforcement

victim/harm-causer and court

court and victim/ harm causer

government

government

government

harm-causer

Updating and Refinement

victim/harm-causer

victim/ harm-causer

government

government

government and harmcauser

Harm-causer

C. Optimal Specificity of Regulation Even when a regulatory body turns out to be the least-cost-information generator, we face a second optimization issue centered on how much particularized data and analysis is it worthwhile to

generate.114 In an ideal world, policies and compensation should be tailored to the individual needs and circumstances to each “victim” and polluting entity. But, specificity comes at an administrative cost. If agencies devote significant resources to tracking and analyzing environmental harms on a case-specific basis, they can produce highly refined and granular policy interventions that reduce the “failure costs” of over- or under-regulation. Alternatively, they can undertake more broad-based analysis, using “average” data and generalized estimates of impacts, harms, and policy gains. As noted earlier, community-scale (or broader) decisionmaking reduces administrative burden. But these analytic shortcuts inflict inefficiencies on society in the form of mis-specified environmental standards.115 The greater the heterogeneity of the polluters as well as the pollutees, the larger will be the likely mis-specification errors—polluters both under-regulated and over-regulated and victims both under-protected and overprotected (or under-compensated or over-compensated). 116 These errors represent the price paid for the reduced information burdens of collectivized environmental decisionmaking. From a “least social cost” perspective, the logic of calculating an optimal specificity of regulation is strong. If the expense of greater regulatory precision cannot be recouped with greater net social benefits, why bother? But the issue is not this simple. Using a welfare maximization calculus to determine the optimal specificity of regulation defines the ends as well as the means of environmental protection. 117

114

Colin Diver, The Optimal Precision of Administrative Rules, 93 YALE L.J. (1983) [hereinafter Diver, Optimal Precision] (setting out the “optimal specificity” of regulation); Isaac Ehrlich & Richard A. Posner, An Economic Analysis of Rulemaking, 3 J. LEGAL STUD 257, 261 (1974); see also RICHARD POSNER, A N ECONOMIC A NALYSIS OF LAW 367-70 (4th ed. 1992) [hereinafter POSNER, ECONOMIC A NALYSIS ] (arguing that the key question is whether the benefits of particularization outweigh the costs); Richard Epstein, SIMPLE RULES FOR A COMPLEX W ORLD 30-36 (1995) (arguing for simpler rules). But see Louis Kaplow, A Model of the Optimal Complexity of Legal Rules, 11 J. L. ECON. & ORG. 150, 161 (1995) (presenting a more sophisticated model for determining how complex rules should be). “Optimal specificity” issues will be discussed in greater detail below. 115

By mis -specified standards, I mean that the marginal costs and benefits of intervention are not carefully calibrated and equilibrated. 116

Robert Mendelsohn, Regulating Heterogeneous Emissions, 13 J. ENVTL. ECON. & M GMT . 301 (1986) (demonstrating that greater heterogeneity leads to higher welfare losses from standardized policies). 117

Many different “least cost” models have been developed. Rose, for example, suggests tracking (1) administrative costs, (2) user costs, and (3) overuse or failure costs. Carol M. Rose, Rethinking Environmental Controls: Management Strategies for Common Resources 1991 DUKE L.J. at 12 [hereinafter Rose, Rethinking Environmental Controls]; see also

40

First, such an approach creates a tension between two dimensions of efficiency. Under a least social cost approach to pollution abatement, whenever the costs of intervention (A + T) exceed the gains from intervention (reduced I), no action will be undertaken. 118 This “let the harms fall as they may” choice maximizes social welfare but may leave externalities uninternalized and thus fail to achieve allocative efficiency in the use of the resources in questions. Social welfare maximization does not therefore translate into full cost internalization but rather an optimal degree of cost internalization from a utilitarian perspective. Second, if precision is sacrificed to reduce administrative burdens, some parties will be stuck with costs they would not otherwise have had to bear. For utilitarians, these mis -specification errors are not an issue. But if we qualify our utilitarianism and the welfare maximization it implies with a corrective justice principle which says that innocent pollutees have a right not to be subject to uncompensated damages (and that regulated entities should not be charged with pollution control burdens that exceed the harm they are causing), we have a justice issue to address. As a general matter, under a system based on gross averages, those who face high levels of harm will be underprotected or under-compensated. From a producer standpoint, facilities with below average emissions will tend to be over-regulated or burdened with higher-than-appropriate pollution control costs. If we have neither a priori assumptions about the fairness of letting pollution harms fall as they may nor other justice considerations to factor into the analysis, the efficiency of a least-social-cost outcome makes sense. But what if we do care about who bears the harms? For instance, what if we have a preference for protecting environmental property rights as well as other property rights.119 Might we not want to adhere

A.B. Jaffe et al., Environmental Regulation and the Competitiveness of U.S. Manufacturing: What Does the Evidence Tell Us? J. ECON. LIT . Mar., 1995, at B2-I63 (noting that the total costs of environmental regulations can be subdivided into administrative costs, compliance costs, and transaction costs); Robert F. Blomquist, Some Mostly Theoretical and (Very Brief) Pragmatic Observations On Environmental Alternative Dispute Resolution in America, 34 VAL. U. L. REV . 343, 357 & n. 29 (2000) (noting that transaction costs consists of several potentially discrete costs including information costs, opportunity costs, and administrative costs). 118

Rose, Rethinking, supra note 117(spelling out the logic of a “do nothing” response in some circumstances).

119

Carol Rose, Property as the Keystone Right, 71 NOTRE DAME L. REV. 329 (1996) (arguing that the protection of property rights is fundamental to a well functioning society).

41

to a principle that privileges the right of breathers to be free from air pollution? 120 Insofar as a leastsocial-cost regulatory approach and economization on information results in externalities not being internalized, we may end up with uncompensated respiratory harms falling on innocent pollutees. Third, mis-specification errors are not likely to be evenly distributed. Those best positioned to provide the information needed to reduce these errors have an incentive to generate or come forward with information only when they will benefit from greater precision. 121 To the extent producers (polluters) are much better positioned to generate data on environmental harms than pollutees, mis-specification errors are much more likely to result in under-regulation rather than over-regulation. D. Corrective Justice If pollution harms and the economic gains from the activities that generate them were distributed evenly across all citizens, one could argue that where the gains exceed the costs, the economic benefits of polluting activit ies justify the damages suffered. In this case, the pollution burden is akin to a tax — the price the community has to pay for economic activity. But if the harms fall disproportionately on a subset of citizens, the pollution burden appears less like a tax and more like a taking, where some individuals have their property seized to fund a community benefit.122 The trade-off between efficiency and corrective justice in the pollution context thus requires more attention. 123 Take, for example, a town with a thousand residents and a polluting factory. If each resident

120

The idea that people should not be forced to bear uncompensated injuries from others and that victims of such harms should be made whole by the harm causer runs deep in the law. See, e.g., DAN B. DOBBS, LAW OF REMEDIES (2d ed. 1993). 121

More broadly, as I discuss in Part V. F. below, the interaction of significant information gaps and a least-social cost regulatory calculus creates a large category of cases where doing nothing seems rational. This bias is reinforced by anecdotal evidence of cases where society appears to have “over paid” for environmental protection. See, e.g., PHILIP K. HOWARD, THE DEATH OF COMMON SENSE (1995). 122

This “taking” could be corrected by compensation payments to those disproportionately affected by the environmental harm. For example, those living near an incinerator could be given regular “pollution victim payments .” But, in practice, this is rarely done. 123

Emphasis on internalizing externalities in a least-social-cost manner translates into a utilitarian approach to policymaking that tends to crowd out corrective justice as a policy goal. Calabresi and Melamed, for example, dismiss justice concerns as either an element of efficiency based on protecting expectations, or a distributional issue that should be separately addressed. Calabresi & Melamed, Cathedral, supra note 3, at 1105, 1123. In elevating efficiency above other

42

suffers pollution harms of $1, we would like to do two things by means of governmental intervention. First, for the sake of allocative efficiency, we would like to internalize the pollution externality by making the factory pay for the $1000 worth of harm it is causing. Second, for the sake of corrective justice, we would like to make the pollution victims whole by giving them each $1. But reality deviates from this scenario in two important ways. First, under command and control regulation, the factory faces government requirements (either technology mandates or performance standards) to reduce emissions by some amount. If we have good regulators, emissions are reduced to the point where the marginal costs of further regulation would exceed the marginal benefits. But this likely leaves emissions at some level above zero, imposing residual pollution harms of some level on the residents. Thus, we will have achieved a “least-cost” outcome but not fully internalized the externality (given ongoing residual emissions) nor made the victims whole. The goals of allocative efficiency and corrective justice have not been met. Alternatively, under a market-based pollution control regime, the regulators might charge the company for its emissions. These pollution fees would likely induce the company to reduce its emissions to some degree (presumably to the point where the marginal cost of additional controls would exceed the marginal benefit in reduced fees). The company would then pay for its residual emissions and the harm they cause. This would fully internalize the pollution externality and produce an allocatively efficient outcome. But the pollution fees would likely be kept by state or federal authorities so the victims would not be made whole. In the real world, moreover, not every citizen is similarly situated. A more realistic scenario would entail 10 citizens (perhaps the next-door neighbors) facing harms of $20 each, a second tier of 100

goals, law and economics scholarship often underattends to the protection of property rights and the need to make victims whole. Calabresi recognized this point but is still better remembered for his efficiency thrust. Calabresi, Pointlessness, supra , note 2, at 1224-28 (noting that distributional analysis becomes inevitable and hence essential); see also Frank Michaelman, Pollution as a Tort: A Non-Existential Perspective on Calabresi’s Costs, 80 YALE L.J. 647 (1971); see also JULES COLEMAN, M ARKETS, M ORALS AND THE LAW (1988). A full discussion about “taking environmental rights seriously” will have to be deferred to another day and another article. But note that the cramped conception of justice offered by Calabresi and others seems inconsistent with an insistence that liability rules cannot be applied to crimes against property or bodily integrity. Cathedral at 1124. Why don’t pollution “assaults” fall into this category?

43

citizens (nearby neighbors) facing $3 each in harms — and the rest of the community suffering damages of just 50¢ each. In this case, the $1000 in pollution fees, if absorbed by the government as a tax offset, dramatically under-compensates the next-door neighbors who receive $1 (in reduced taxes) but suffer harms of $20 somewhat under-compensates the nearby neighbors who face harms of $3 but receive just $1, and over-compensates everyone else as they receive $1 but faced harms worth only 50¢. Thus, under a scenario where the harms are not spread equally, even an allocatively efficient pollution control regime fails to meet the corrective justice goal of making victims whole. In light of the high probability that real-world environmental policymaking fails to compensate victims in a fashion commensurate with their harms, why is the corrective justice dimension of environmental law and policy so often neglected? 124 Several reasons can be identified. First, distributional considerations may cut the other way. If those who bear disproportionate environmental damages are the most affluent members of the community then the structure of a pollution “tax” as the price for economic activity may help to achieve greater equality. But as an empirical matter, it seems likely that the equity argument goes the other way — and that those who are undercompensated for pollution harms will be poorer than the average.125 Second, some observers see the disregard for pollution harms as a policy choice to promote industrialization and economic growth.126 Other scholars argue that the pattern of neglect for pollution victims took hold in the 19th Century as a reflection of the power of the emerging industrial class, which constructed a legal regime that favored their economic

124

It might be argued that this neglect is ending as the “environmental justice” movement has emerged. See generally, Richard L. Lazarus, Pursuing Environmental Justice: The Distributional Effects of Environmental Protection, 87 N. W. U. L. REV. 787 (1993). 125

Vicki Been, Locally Undesirable Land Uses in Minority Neighborhoods: Disproportionate Siting or Market Dynamics, 103 YALE L. J. 1383 (1994) (explaining that market forces mean environmental harm disproportionately fall on poorer people); William Bowen, An Analytic Review of Environmental Justice Research: What do we Really Know? 29 ENVT ’L. M AN. 321 (2002) (surveying numerous studies showing poverty connection to disproportionate environmental harms). 126

M ORTON J. HOROWITZ, THE TRANSFORMATION OF A MERICAN LAW , 1780-1860 (1977) (discussing neglect for pollution as a subsidy to industrialization); see also Herbert Hovenkamp, The Economics of Legal History, 67 M INN. L. REV. 645, 670-78 (1983).

44

interests.127 Still others conclude that justice considerations dropped out of the environmental equation as a result of utilitarian legal analysis 128 or a judicial desire to promote wealth maximization. 129 While all of these arguments are plausible and each represents part of the explanation for the minimal role corrective justice plays in our present day pollution control regime, I think another reason must be considered: the historically high cost of protecting environmental property rights on an individualized basis. Sorting out who has been harmed and how much they should be compensated has been considered to be too information intensive. In my example, if the residual harm were judged to be $1 per resident and the cost of delivering the compensation to the 1000 residents were greater than $1000, there would seem to be no point in going forward. The costs of intervention would exceed the benefits. This decision must be understood, however, as a choice to elevate welfare maximization over the protection of property rights and justice. Now consider how the picture changes if the cost of delivering compensation falls from $1000+ to $1. Might we not then want to make the victims whole? And what if we were able to measure which residents were harmed and by how much — and could thus tailor the compensation payments to the precise level of harm each pollutee suffered? Would we not then want to provide collective justice alongside efficiency? To the extent that the digital technologies lower the cost of analysis and other transaction costs, they may also revive interest in corrective justice by softening the apparent trade-off with economic efficiency. The ends as well as the means of environmental policymaking may therefore evolve in the Information Age.

127

PAUL H. RUBIN, BUSINESS FIRMS AND THE COMMON LAW : THE EVOLUTION OF EFFICIENT RULES , 3-30 (1983); Christine Rosen, Differing Perceptions of the Value of Pollution Abatement Across Time and Place: Balancing Doctrine in Pollution Nuisance Law, 1840-1906, 11 LAW & HIST . REV. 303, 354-55 (1993) [hereinafter Rosen, Differing Perceptions]. 128

Rosen, Differing Perceptions, supra (arguing that utilitarian analysis has resulted in under-attention to pollution).

129

Richard A. Posner, Wealth Maximization and Judicial Decision-Making, 4 INT ’L. REV. L. ECON. 131 (1984); Richard A. Posner, A Theory of Negligence, 1 J. LEGAL STUD. 29 (1972) (arguing that judges have pursued efficiency goals even before the economic logic was fully understood).

45

IV. INFORMATION AGE OPPORTUNITIES FOR ENVIRONMENTAL P ROGRESS While a frictionless Coasean marketplace in environmental property rights may not be around the corner, the rapid evolution and diffusion of digital technologies represents a “supply shock” in the environmental information marketplace with broad implications. This Part spells out the technological logic for expecting a future environmental world of much lower transaction costs—where we can more readily identify each unit of pollution harm, place a value on it no matter how small, trace that unit to its source, and charge for (or regulate) the harms caused. A. Bringing the Information Age to Bear on Environmental Protection The technological opportunities presented by the Information Age have been the subject of a vast amount of writing, both popular and scholarly. 130 Several Information Age breakthroughs promise to transform the environmental problem set and to reconfigure our portfolio of response strategies to pollution control and natural resource management challenges. 1. Computer Power At the heart of the Digital Revolution lies the computer, which makes the job of gathering, sorting, storing, and analyzing data much faster, easier, and cheaper.131 Environmental problem solving has been defined by the need to simplify the problems presented and to overcome omnipresent uncertainties. Without eliminating all of the uncertainties, computers make it possible to fill data gaps in a systematic and sophisticated manner and to better respond to problems that involve multiple variables. Computer-generated gains in information management will encourage a more empirical and data-driven approach to environmental protection. The ability to track and use vast quantities of data promises to

130

M ICHAEL L. DERTOUZOS , W HAT W ILL BE: HOW THE NEW W ORLD OF INFORMATION W ILL CHANGE OUR LIVES (1998) (unveiling the many upcoming technologies of the Information Marketplace, including the personal television, Internet, and phone networks, or “bodynets,” that will drastically transform daily life); GATES , THE ROAD A HEAD, supra note 8 (offering his vision as to how Information Age technologies will affect education, business, politics, commerce and the home); KEVIN KELLY, OUT OF CONTROL: THE NEW BIOLOGY OF M ACHINES , SOCIAL SYSTEM AND THE ECONOMIC W ORLD (1998); STAN DAVIS & CHRISTOPHER M EYER, BLUR : THE SPEED OF CHANGE IN THE CONNECTED ECONOMY (1998); Power to the Mobile People, W IRED, Apr. 12, 2003 [hereinafter, DAVIS & M EYER, BLUR ] (predicting how the interface of dimesized fuel cells and mobile technologies will soon allow for unlimited connectivity). 131

NICHOLAS NEGROPONTE, BEING DIGITAL (1996) (reviewing Digital Age breakthroughs).

46

change the optimal specificity of regulation, permit more complex and individualized policy interventions, and make diversity easier to manage. Computer-led advances in other fields, such as epidemiology, hydrology, and statistical modeling, are reshaping the foundations on which environmental decisionmaking builds.132 The sheer computational power of today’s computers makes possible analytic endeavors that were unimaginable just a few years ago.133 The boundaries of learning are thus being pushed out at an ever accelerating rate.134 Computers are also bringing efficiency gains to every corner of the economy, some of which will translate into improvements in resource productivity and thus reduced pollution. 135 Across the spectrum of information needs identified in Table 1, Information Age breakthroughs promise to make it easier to fill the gaps that plague environmental decisionmaking. 2. Communications Communications breakthroughs represent a second dimension of the Information Age with potentially important environmental implications. From cell phones to palm-sized personal digital assistants to global positioning systems, technologies have emerged which portend not only the “end of

132

For example, the number-crunching capacity of super computers dramatically improves the human ability to model critical systems and to forecast results. In the environmental domain, analysts can now track the dispersion of pollutants with much greater precision than ever before. Similar gains in statistical methods (including advanced regression and “neural net” analysis), econometric modeling, epidemiological analysis, cost-benefit forecasting, and risk analysis all promise to dramatically improve the quality of the underlying information base on which environmental policy decisions are made. See, e.g., ROBYN N. DAWES , EVERYDAY IRRATIONALITY: HOW PSEUDO-SCIENTISTS , LUNATICS, AND THE REST OF US SYSTEMATICALLY FAIL TO THINK RATIONALLY (2001) [hereinafter DAWES , IRRATIONALITY] (demonstrating that decisions based on data and statistical prediction almost always outperform those based on expert judgment). 133

BRADLEY EFRON, THE JACKKNIFE, THE BOOTSTRAP , AND OTHER RESAMPLING PLANS (2002) (highlighting the gains made possible by expanded computing power). 134

Robert Wright, Mr. Order Meets Mr. Chaos, 124 FOREIGN POL’Y . 50 (May/June 2001) (forecasting an optimistic future through increasing technological innovation); but see THOMAS HOMER-DIXON, THE INGENUITY GAP (2000) (arguing that problems are emerging faster than solutions). 135

Salzman, Beyond the Smokestack , supra note 43; Daniel C. Esty, Digital Earth: Saving the Environment, OECD OBSERVER, May, 2001, at 68. Some observers fear that a booming Digital economy will lead to more pollution as electricity demand spikes and ever-more-quickly-outdated electronic devices, full of heavy metals and toxics, must be disposed of. Fear of a “rebound effect” seems overstated. Efficiency gains seem likely to exceed scale effects. Gene Grossman & Alan B. Krueger, Economic Growth and the Environment, 110 Q. J. ECON. 353 (1995). And recycling of old computers and cell phones is a booming business. William L. Thomas, Rio’s Unfinished Business: American Enterprise and the Journey Toward Environmentally Sustainable Globalization, 32 ENVTL L. REP . 10873 (2002) (noting computer and cell phone recycling initiatives in U.S. and Europe).

47

distance” but also the “collapse of time.”136 Connection speeds are increasing at a dramatic pace while the cost of communications is falling rapidly. 137 Links to virtually anyone on the planet or to data sets anywhere in the world at any time at very low cost are now available. This hyper-connectivity changes the cost of establishing an appropriate technical baseline for environmental decisionmaking. Relevant information — on-the-ground facts, answers to many scientific questions and data on how others have dealt with particular issues on problems — will be much more readily available. 138 Just as computers help to generate the information and analysis that makes a more data-driven approach to environmental protection possible, communications systems make the transmission of this information to where it is needed fast, easy, and inexpensive.139 While the advantages of low-cost communications seem substantial, hyper-connectivity may have some environmental disadvantages as Part VII explores in more detail. Increased “network effects” may lock in sub-optimal standards and approaches.140 Better information generally improves decisionmaking, but more information could include more disinformation or propaganda.

136

Time and distance, which both tend to translate into cost, have often meant that potentially useful information is not brought to bear on the environmental realm. M ITCHELL, E-TOPIA, supra note 8 (explaining how information technologies reduce time and distance); DOUGLAS F. A LDRICH, M ASTERING THE DIGITAL M ARKET PLACE 31 (1999) (highlighting time impacts of digital breakthroughs) [hereinafter A LDRICH, M ASTERING]. 137

FIRST 2000 TRENDS IN TELEPHONE SERVICE REPORTS 14-4 (2000) (charting, among other variables, the changes in the prices of directly dialed five-minute long distance calls). 138

More plentiful and less expensive information does not guarantee that unprocessed data will be translated into usable knowledge. Data is the raw material. Information is the intermediate good, reflecting some processing of the data. Knowledge is the final product where analysis allows us to extract conclusions. To move from data to substantive content requires a rigorous scientific process of developing and testing hypotheses, careful analysis, systematic quality controls, and sensitivity analysis to highlight the effects of remaining uncertainties. A LDRICH, M ASTERING, supra note 136, at 224. 139

Daniel C. Esty, Toward Data-Driven Environmentalism: The Environmental Sustainability Index, 31 ENVTL. L. REP . 10603 (2001). 140

M ALCOLM GLADWELL, THE TIPPING POINT : HOW LITTLE THINGS CAN M AKE A BIG DIFFERENCE (2000); Michael L. Katz & Carl Shapiro, Systems Competition and Network Effects, 8 J. ECON. PERSP., 93 (1994) (discussing network effects).

48

3. Internet At the intersection of the computer and the communications revolution lies the Internet, which provides a systematic method for connecting people and knowledge at very low cost.141 The network structure of the Internet allows access to vast quantities of information that would previously have been unobtainable or unsearchable in a cost-effective manner. In doing so, it expands the pace of knowledge exchange,142 including knowledge about how to address environmental harms.143 The Internet also makes new markets possible.144 By lowering search costs, buyers and sellers who might previously not have found each other now can.145 In effect, the speed and low cost of data access on the World Wide Web makes comparison shopping easy to do and facilitates comparative analysis of many other types. Comparative data turns out to be a powerful tool for better decisionmaking in the environmental domain , which is marked by high degrees of uncertainty. In particular, comparative data permits a shift toward a more empirical and data-driven approach to environmental decisionmaking. The Internet provides low cost and easy access to vast stores of data which make a quantitative decision process much easier to manage. Indicators that gauge on-the-ground circumstances allow officials to determine which issues are becoming more serious and deserve to be priorities and which are becoming

141

M ITCHELL, E-TOPIA supra note 8 (discussing the time, space, transportation, and coordination efficiencies from digital networks); s ee also , A. Voinov & R. Lostanza, Watershed Management and the Web, 56 J. ENVTL. M GMT . 231-245 (1999) (showing how the Internet can advance and improve watershed management); Dan Taylor, Environment and Infrastru cture. How We Can Use It to Avert Another Tragedy of the Commons, INFORMATION IMPACTS M AGAZINE, Oct. 1999 at http://www.cisp.org/imp/October_99/10_99taylor-insight.htm (proposing an Internet-based management monitoring system for the Great Lakes). 142

Ramamurti Shankar, Globalization and Science: A Speeded-up Virtuous Cycle, Yale Global Online (Mar. 28, 2003), available at http://yaleglobal.yale.edu/display.article?id=1259 - 15.3KB. 143

INFORMATION SYSTEMS AND THE ENVIRONMENT (Deanna J. Richards et al., eds., 2001) (highlighting various ways the Internet is reshaping environmental policymaking). 144

The power of the Internet to create new markets is now a constant refrain. See, e.g., A JIT KAMHIL & ERIC VAN HECK, M AKING M ARKETS : HOW FIRMS CAN DESIGN AND PROFIT FROM ONLINE A UCTIONS AND EXCHANGES (2002); JOHN HAGEL III & ARTHUR G. A RMSTRONG, NET GAIN : EXPANDING M ARKETS THROUGH VIRTUAL COMMUNITIES (1997); ROBERT E. HALL, DIGITAL DEALING: HOW E-M ARKETS ARE TRANSFORMING THE ECONOMY (2002); A DAM COHEN, THE PERFECT STORE: INSIDE EBAY (2002). 145

See, e.g ., www.recycle.net, described as a “worldwide trading site for information related to secondary or recyclable commodities, by-products, used & surplus items or materials.” The Internet will also likely prove instrumental in many of the emergent carbon and other emissions-trading markets developing throughout the globe.

49

less serious and should be downgraded as a focus. Similarly, measurement permits “benchmarking” across jurisdictions, facilitating comparisons of technology and policy results and the identification of superior environmental strategies. Enormous environmental progress could be achieved simply by moving lagging jurisdictions toward the “best practices” of those with top-ranked results. Quantitative measurement and easy access to this data provide a rigorous basis for judging which specific regulatory approaches, technologies, or corporate activities are delivering results and should be replicated—and which are not succeeding and need to be rethought. Comparative analysis, moreover, spurs competition. Evidence that others are outperforming one’s country, community, or company on environmental criteria can heighten attention to opportunities for improved pollution control and better resource management. And competition not only motivates better performance it also unleashes innovation. 146 The Internet furthermore democratizes information access and increases “transparency.” Everyone with a computer has a “window” on government activities and corporate actions as well. 147 Openness encourages broad-based participation in decisionmaking, consideration of a range of viewpoints, and careful sifting of facts and arguments. More thorough and robust debate yields better environmental answers over time.148 Data access across the Internet will not eliminate all uncertainty in the environmental domain. But even where significant questions remain, the availability of data from different perspectives and circumstances will permit a process of triangulation that may help to illuminate the underlying issues and narrow the zone of uncertainty.

146

M ICHAEL E. PORTER, COMPETITIVE A DVANTAGE OF NATIONS (1990) (explaining why competitive pressures generate innovation). 147

This logic only applies o nly in parts of the world where computers are common. The issue of a “Digital Divide” remains significant in some places. See Frances Irwin & Carl Bruch, Information, Public Participation, and Justice, 32 ENVTL. L. REP . 10784 (2002) (arguing that despite some notable efforts to cross the Digital Divide and to create a digital dividend, much of the United States’ investment went into the technology bubble rather than into building a more sustainable world). 148

REGULATORY COMPETITION, supra note 21. The potential downsides of greater transparency are discussed in Part VII.

50

4. Digitalization As digital technologies are deployed across the economy and society, the ripples of the Information Age will extend, affecting environmental policymaking in a number of ways. Remote sensing from satellites in space149 and other new sensor technologies150 will soon provide the capacity to see and understand on-the-ground conditions from anywhere at any time at increasingly low cost. Digital-technology-enabled breakthroughs in laser technologies, ion beam analysis, nanotechnologies, and small-scale sensors are already providing a vastly improved ability to detect and measure emissions and contaminants at a fine-grained level. 151 We are thus approaching the day when virtually all pollutants will be susceptible to tagging, tracking, and measurement at relatively low cost.152 Where sheer numerical quantity might once have overwhelmed analysts, computers permit the sifting of raw data in very efficient and high speed ways. Information can be extracted even under conditions that present very high noise-to-signal ratios.153 Advanced statistical techniques, such as neural nets, allow pattern recognition and thus problem identification in much more sophisticated ways.154 And

149

Anthony Vodacek, Environmental Applications of Remote Sensing, INFORMATIQUE, Aug. 2000, at 21; see also Fred Hansen, Three Emerging Issues, Three Needed Changes, ENVTL. F., May/June 2000, at 46 (noting that remote satellite sensing will enable “identification with pinpoint accuracy of all emission discharges”); Paul F. Uhlir, Applications of Remote Sensing Information in Law: An Overview, in EARTH OBSERVATION SYSTEMS: LEGAL CONSIDERATIONS FOR THE ‘90S 8, 16-17 (1990) (describing how remote sensing systems will lead to better ecological understanding and management). 150

The scientific literature is rich with examples of breakthrough in sensors and monitoring technologies. See, e.g., E.J. Hinsta et al., SPADE H2O Measurements and the Seasonal Cycle of Stratospheric Water Vapor, 21 GEOPHYSICAL RESEARCH LETTERS 2559 (1994); C.R. Webster et al., Quantum Cascade Laser Measurements of Stratospheric Methane (CH4) and Nitrous Oxide (N2O), in A PPLIED OPTICS (2001). 151

BLUR , supra note 130 (discussing sensors in cars). Some observers even forecast a time when humans could be outfitted with a “digital skin” that allows a precise record of pollution impacts to be generated. A. Scott Matthews, The Environmental Implications of the Growth of the Information and Technology Sector, at 34 (unpublished manuscript, presented 16 March 2001 at OECD Environment Directorate Workshop, Paris). 152

Allenby et al., Overview and Perspectives, in INFORMATION SYSTEMS AND THE ENVIRONMENT (Deanna J. Richards et al. eds., 2001) (spelling out pollution tracking potential). 153

M. GOEBEL & L. GRUENWALD, A SURVEY OF DATA M INING AND KNOWLEDGE DISCOVERY SOFTWARE TOOLS ; SIGKDD EXPLORATIONS : NEWSLETTER OF THE ACM SPECIAL INTEREST GROUP ON KNOWLEDGE DISCOVERY AND DATA M INING 1, 20-33 (1999); TREVOR HASTIE, ROBERT TIBSHIRANI, & JEROME FRIEDMAN, THE ELEMENTS OF STATISTICAL LEARNINGDATA M INING, INFERENCE, AND PREDICTION (2001) (describing the application of modern statistical techniques for sorting data). 154

SERGIOS THEODORIDIS & KONSTANTINOS KOUTROUMBAS , PATTERN RECOGNITION (1999) (explaining neural networks,

51

genetic algorithms make it possible for decisionmakers to model and test policy options and combinations of options in hours rather than over years. By tightening feedback loops, the evolutionary process of policy refinement through trial and error can be dramatically sped up. 155 Improved modeling techniques and broader data foundations will greatly facilitate fate and transport analysis and the tracking of pollution effects. The same technologies, applied in the fields of ecological science and epidemiology, promise to improve our understanding of dose-response relationships, cumulative exposures, and interactions that aggravate or mitigate environmental risks. Better data and enhanced analytic tools will also strengthen our capacity to make economic decisions, carry out cost-benefit studies, and make harm valuations — broadly reducing the panoply of environmental information gaps identified in Part II.

V. RETHINKING ENVIRONMENTAL P ROTECTION IN AN INFORMATION-RICH WORLD As described above, the advances of the Information Age have shifted our environmental protection “possibility frontier” and opened the door to a new era of pollution control and natural resource management.156 More data, better analytic tools, and fewer information gaps make improved performance possible across the existing spectrum of institutional approaches to pollution control and natural resource management. The new technologies will also restructure the optimal mix of policy tools and institutions. At the same time, new policy challenges may be created and some pre-existing regulatory pathologies may be exacerbated.

rule learning systems and statistical analysis as classifiers or clusterers). 155

Learning as the key to a process of environmental progress through trial and error has been stressed by a number of commentators. DANIEL A. FARBER, ECO-PRAGMATISM: M AKING SENSIBLE ENVIRONMENTAL DECISIONS IN AN UNCERTAIN W ORLD (1999); J. B. Ruhl, James Salzman, & Kai Sheng Song, Regulatory Traffic Jams, 2 W Y. L. REV. 253 (2002); KRIER & URSIN, POLLUTION AND POLICY, supra note 52. 156

In discussing the “pointlessness” of Pareto Optimality, Calabresi urges a focus on eliminating obstacles to efficiency rather than hoping for a shift in the Pareto frontier. Calabresi argues that the latter strategy is no better than waiting for manna to fall from the Heavens. While manna may not be falling, the cost of information is, effecting just the sort of technology shift in the Pareto frontier Calabresi dismissed. Calabresi, Pointlessness, supra note 2 , at 1218-1219.

52

A. Revitalizing the Environmental Property Rights Market For more than a century, as environmental problems became more complex and the resulting information demands led to widespread market failure, reliance on a property rights approach to environmental protection has been in decline. New digital technologies seem likely to reverse this trend by reducing several types of information failures that now limit the use of Coasean bargaining in the environmental context. 1. Search Costs One of the most striking features of the Internet is its capacity to “make markets” and to improve the efficiency of existing systems of exchange.157 By lowering search costs, the Internet brings buyers and sellers together who might otherwise not have found each other. The market-making capacity at the Internet has various environmental applications. Most notably, more “waste” can be re-used and recycled. The re-use opportunity emerges where one company’s waste or by-products serve as another’s raw material. 158 Historically the cost to the waste generator of finding a synergistic partner would often have been unduly high. As a result, byproducts were discarded. With the Internet, matchmaking costs go down dramatically. In fact, Internet-driven reuse/recycling relationships are multiplying at a very rapid pace.159 The chemical maker Rhone-Poulenc found, for instance, that the diacids which are a byproduct of nylon production need not be incinerated as waste but can be sold to tanneries for use as coagulants.160

157

KEVIN KELLY, OUT OF CONTROL: THE NEW BIOLOGY OF M ACHINES , SOCIAL SYSTEMS, AND THE ECONOMIC W ORLD 184199 (1994); GATES , THE ROAD A HEAD, supra note 8. 158

Marian Chertow, Industrial Symbiosis: Literature and Taxonomy, 25 A M. REV. ENERGY & ENV’T 313 (2000); Keith Pezzoli, Environmental Management Systems (EMSS) and Regulatory Information, 36 CAL. W.L. REV. 335, 364 & n. 187 (2000) (providing examples of industrial symbiosis). 159

For a list of more than a hundred Internet bulletin board systems that are facilitating business-to-business “waste” transfers, see http://www.recycle.net/recycle/exch and http://www.wasteexchange.org/exchanges. 160

Michael E. Porter & Claas van der Linde, Green and Competitive: Ending the Stalemate, 73 HARVARD BUS . REV. 120, 125 (1995).

53

Along with making new markets, the Internet, e-commerce, and more plentiful information generally promise to create better informed consumers and thus to transform existing markets. Because the cost of getting information on the environmental dimensions of products is now high, purchasers usually ignore environmental factors. But as environmental data and analysis becomes more accessible, additional buyers are factoring environmental considerations into their choices. Eco-labels and other tools have created a wave of “Green consumers” in a number of contexts from electric power to furniture to food. 161 2. Scope of Environmental Rights Coasian bargaining often seems untenable because those holding environmental rights are in no position to bargain over their exchange. They lack information on a range of issues that would be required for a functioning market. But easier access to “technical” information could facilitate the revival of a property-rights-based environmental protection regime. As pollution harms become easier to track, measure, and “price,” individuals, communities, and other rights holders will be better positioned to understand the harms they face, to make causal connections from emissions to effects, and to place a value on environmental injuries, thus narrowing the information gaps that now limit the viability of a system based on the exchange of entitlements. Environmental databases have historically been patchy and unreliable —woefully inadequate to the needs of efficient decisionmaking, never mind negotiation. However, signific ant strides are now being made in the realm of environmental indicators, metrics, and measurement.162 Air emissions are already capable of being “tagged” and tracked downwind. 163 Advanced meteorological models have 161

Already, a number of entities are on-line seeking to fill this environmental information market niche. See e.g., ENVIRONMENTAL RESOURCES INDEX, ENVIRONMENTALLY FRIENDLY PRODUCTS at http://wwwziplink.net/~cridgway/environ/products.htm (last visited Nov. 5, 2000). 162

M ATHIS W ACKERNAGEL & W ILLIAM REES , OUR ECOLOGICAL FOOTPRINT (1996); OECD, TOWARDS SUSTAINABLE DEVELOPMENT : ENVIRONMENTAL INDICATORS (1998); Press Release, UNEP, Global Study on State of the World’s Ecosystems Launched to Mark World Environment Day (June 5, 2001); see also W ORLD ECONOMIC FORUM, 2002 ENVIRONMENTAL SUSTAINABILITY INDEX (2002) available at www.yale.edu/envirocenter/esi. 163

EPA has developed a variety of new technologies to track air emissions. Darla Carter, New Wind Sensor Gets Test Run, COURIER J., June 24, 2000, at 1B.

54

further strengthened our ability to understand the sources and “receptors” of air pollution. Indeed, Gaussian plume analysis now permits regulators to plot the drift of air pollutants with a degree of precision unheard of just a few years ago.164 And Internet-based tools such as the emissions “scorecard” (www.scorecard.org) created by Environmental Defense increasingly put this data at everyone’s fingertips. Similar gains are being made in tracking pollutants as they flow through watersheds and hydrological systems.165 It may not be all that many years before emissions sources will be nearly completely mapped if not fully understood. 166 As a leading toxicologist observes: Admittedly, there are a large number of factors to consider when estimating exposure, and it is a complicated procedure to understand the transport and distribution of a chemical that has been released into the environment. Nonetheless, the available data indicate that scientists can do an adequate job of quantifying the concentration of the chemicals in the various media and the resulting uptake by exposed persons. . . 167 Comparable gains are being made in our understanding of environmental public health effects and the ecological impacts of various types of pollution. In combination with the extraordinary advances in understanding the sensitivity of the human body to various kinds of chemical exposures, we stand on the verge of much more scientifically defensible answers to questions about how much harm certain 164

Such modeling has transformed the debate over the sources of air pollution in the Northeastern United States. For an example of ozone modeling, see the final report and technical supporting document produced by the Ozone Transport Assessment Group, at http://www.epa.gov/ttn/lto/otag/finalrpt (last visited Apr. 5, 2001). 165

Kurt Stephenson et al., Effluent Allowance Trading: The Nonpoint Source Challenge, 16 CONTEMP. ECON. POL'Y, 412, 415-418 (1998) (describing several programs successfully measuring the movement of non-point source pollutants through a system); but see Rena Steinzor, Devolution and the Public Health, 24 HARV. ENVT ’L. L. REV. 351 (2000) (rejecting the possibility that we can know enough about harms to make trading worthwhile). 166

Dennis J. Paustenbach, The Practice of Exposure Assessment: A State-of-the-Art Review, 3 J. TOXICOLOGY & ENVTL. HEALTH 179 (2000) [hereinafter Paustenbach, Exposure Assessment] (chronicling extraordinary recent gains in understanding environmental risk exposures); see also Jennifer Brown, Pediatric Environmental Health Hazards and the Role of Government in Adopting Standards to Protect Children, 16 PACE ENVTL. L. REV. 189, 197-202 (1998) (describing new national initiatives and research into the effects of pediatric exposure to pollutants); Carl B. Meyer, The Environmental Fate of Toxic Wastes, The Certainty of Harm, Toxic Torts, and Toxic Regulation, 19 ENVTL. L. 321, 32224 (1998) (citing advances in detecting toxic exposure: “Toxic wastes can be fingerprinted; emission sources can be tracked with the toxic residues found in the bodies of environmental injury victims; the extent of accumulation of toxics in the environment can be predicted; and the body burden of toxic exposure victims can be measured with a high degree of scientific certainty.”) (Citations omitted). 167

Paustenbach, id. at 264.

55

pollutants cause to particular individuals. The sweeping potential for improved causal specification derives not only from new environmental technologies but also the knowledge revolution within other fields including medicine, public health, epidemiology, and risk-benefit analysis.168 This narrowing of the band of uncertainty around problem identification, the fate and transport of emissions, harm evaluation, and damage valuation, removes a major obstacle to the pricing of pollution externalities and thus the functioning of environmental property rights markets.169 3. Property Demarcation Sophisticated information management systems can often reduce the cost of delineating environmental rights. 170 For example, Global Positioning Systems (GPS) and Geographic Information Systems (GIS) make the physical demarcation of property boundaries faster, easier, and cheaper.171 These technologies, in combination with improved sensors and advances in scientific understanding of how various pollutants move through the air and water or across land, promise to narrow the range of disputes over environmental property rights intrusions. Knowing who has been harmed by what and to

168

Researchers at the Harvard School of Public Health, for example, are using advanced computer modeling to better assess the health impacts of air pollution. Study Details Impact of Pollution on Public Health from Nine Older Fossil Fuel Power Plants in Illinois, A SCRIBE NEWSWIRE, Jan. 3, 2001. Genetic research techniques are also being applied to investigate the effects of various toxics on humans and other organisms. Deborah Schoen, A Renaissance for Genotoxicity Testing?, 11/1/98 ENVTL. SCIENCE & TECH. A498 (1998). 169

The EPA provides results of its annual Toxic Release Inventory online in very accessible formats, including state fact sheets, data release reports, state date files, and searchable databases. Toxic Release Inventory: Community Right-toKnow, EPA at http://www.epa.gov/tri/7 (last visited Jan. 20, 2000). Advocacy groups provide similarly user-friendly environmental information. The Environmental Defense Fund interactive Scorecard site invites visitors to “find your community — just enter your zip code and find out what pollutants are being released into your community — and who is responsible.” Available at www.scorecard.org (last visited May 2, 2003). 170

This potential has been recognized for some time. GARY LIBECAP , CONTRACTING FOR PROPERTY RIGHTS 17 (1989) (“New technology, which lowers the costs of delimiting individual claims, detecting rule violations, arbitrating disputes, and punishing violators, provides for further gains from applying a more specific assignment of property rights to reduce common pool losses.”). A host of new technologies are currently being developed and applied, including remote sensing of individual car emissions. See Todd Hartman, Plan Aims to End Emissions Test: Remote Screening Would Target Polluters, But Clean Vehicles Would Get Pass in Mail, ROCKY MTN. NEWS , Aug. 31. 2000, at A4. 171

Scott D. Makar & Michael R. Makar, Jr., Geographic Information Systems: Legal and Policy Implications, 69 FLA. BAR J. 44 (1995) (providing an introduction to GIS, its uses, and related legal issues).

56

what extent will reduce a major element of transaction costs leading to market failure and thus regulation. 172 In combination with remote sensing, GPS and GIS technologie s offer particular potential with regard to the management of various open access natural resources. Problems of the “commons,” such as how to prevent over-exploitation of the fish in the ocean, have long vexed environmentalists. 173 The fact that some shared resources are difficult or impossible to police creates a potential for suboptimal results as each individual maximizes his or her own consumption at the expense of the community as a whole.174 But information technologies can be deployed as “virtual barbed wire,” permitting the fencing off of shares in common resources that might have been seen as ineluctably “open access” even a few years ago. For example, fish quotas can be allocated and enforced using satellite tracking to observe and even measure the number of fish being taken by particular vessels. And while the prospect of keeping track of hundreds or even thousands of fishing boats might have seen daunting a few years ago, today’s computers can manage the task with relative ease.175 In fact, New Zealand’s fisheries have recently been revived under a tradeable quota regime reinforced by a sophisticated electronic tracking system. 176

172

Likewise, clearer property rights, which are easier to vindicate, may also precipitate a shift from liability rules toward property rules. 173

Frederick W. Bell, Technological Externalities and Common-Property Resources: An Empirical Study of the U.S. Northern Lobster Fishery, J. OF POL. ECON. Jan.-Feb. 1972, at 148-58 (documenting overexploitation of a commons system fishery), cited in Donald R. Leal, Homesteading the Oceans: The Case for Property Rights in U.S. Fisheries, PERC POLICY SERIES , Aug. 2000, at 3. 174

Garrett Hardin, The Tragedy of the Commons, 168 SCI. 1243 (1968) (explaining the over-exploitation dynamic).

175

Satellite tracking of fishing fleets is now both well-developed and cheap enough that it is marketed directly to private corporations in the marine industry for individual fleet communication and observation. SASCO, Economical Satellite Communications and Tracking for the Marine Industry at http://www.sasco-inc.com/defeult.htm (last visited Nov. 7, 2000). Satellite tracking is also being incorporated formally into environmental regulation in Britain. Statutory Instrument 2000 No. 1078 (W.71). The Sea Fishing (Enforcement of Community Satellite Monitoring Measures) (Wales) Order 2000. 176

Michael De Alessi, Fishing for Solutions, IEA STUD.

ON

ENV’T ., Nov. 11, 1998, at 40-43.

57

In light of such breakthroughs, our view of what constitutes an inherently collective resource may well shrink dramatic ally. Allocation of shares, even in the atmosphere, may become possible.177 Moreover, where resources are, by choice, managed as a public good (e.g., parks) the capacity to enforce limits on exploitation of the resource by individual actors will be greatly enhanced. From electronic signup sheets for tent sites in Yellowstone National Park to the tracking of greenhouse gas emissions released into the earth’s atmosphere, the costs of defining and protecting, as well as trading, environmental property rights appear likely to drop precipitously in the coming decades. Information advances may also ease the strain that often exists where resources are shared between private property owners and public entities. Computer-aided mapping and data storage in highresolution geographic information systems makes the presence of endangered species or wetlands much easier to predict. Predictability, in turn, reduces the chances that those buying land will face a userestricting “surprise” that raises a takings issue.178 Simultaneously, the extent of “positive externalities” generated by private landowners will be much easier to gauge, facilitating compensation or credit for those whose careful private land management practices generate public benefits from local open space to biodiversity in the Amazonian rain forest.179 4. Valuation and Strategic Behavior Even if we know with a reasonable degree of precision the sources and impacts of emissions and who holds the relevant property rights, a market-based approach to environmental protection remains untenable unless those engaged in exchanges know how much value to place on the various resources or harms in question. Important dimensions of the “valuation” problem are, however, amenable to Information Age advances. New technologies and firmer data foundations allow cost-benefit analysis to 177

Jonathan Baert Wiener, Global Environmental Regulation: Instrument Choice in Legal Context, 108 YALE L. J. 677, 709-713, 763-768 (1999). 178

To the extent that the Supreme Court’s jurisprudence in this arena turns on “reasonable investment-backed expectations,” increased information should reduce over time the number of ESA takings cases and wetlands disputes to near zero. See Lucas v. South Carolina Coastal Council, 505 U.S. 1003, 1017 (1992). 179

Carol Rose, Property Rights and Responsibilities, in THINKING ECOLOGICALLY , supra note 44, at 56-7 (arguing for expanded compensation to property owners).

58

become more sophisticated and the resulting valuations more refined. Similarly, as analysts become better able to track environmental threats, understand more clearly the probabilities of harm, and forecast the magnitude of ill effects, much greater precision in risk-benefit calculations becomes possible .180 All of this makes a more empirical approach to environmental protection possible. Indeed, a shift toward data-driven decisionmaking and comparative analysis has recently emerged in the environmental realm.181 An “accounting” approach to environmental protection will not eliminate information gaps and concomitant decisionmaking breakdowns, bargaining failures, and strategic behavior. But better data, in combination with a more transparent policymaking process, makes anomalous value claims stand out more sharply. For example, in trying to place a value on a child’s pollution-induced asthma the amounts paid in compensation in other similar cases will be easily accessible. Thus, a rights holder who tries to overplay his hand and advance an inexplicably large claim will be more easily stymied. Value judgments will, of course, still have to be made.182 The price to be placed on particular kinds of harms, such as loss of a life, an increased risk of cancer, or a diminished view are not obvious. These questions are not so much technical as they are philosophical and political. 183 But a more information-rich decisionmaking process can help to achieve agreement on the aspects of a problem that are factual, reduce the scope for dispute, and highlight past patterns of compensation. The assumptions

180

James K. Hammit, Data, Risk, and Science, in THINKING ECOLOGICALLY , supra note 44 (highlighting opportunities for better benefit-cost analysis); CASS R. SUNSTEIN, RISK AND REASON: SAFETY, LAW , AND THE ENVIRONMENT (2003) (reviewing various ways to improve the foundations of environmental analysis ); ROBERT W. HAHN, REVIVING REGULATORY REFORM: A GLOBAL PERSPECTIVE (2001) (discussing possible improvements in cost-benefit analysis). 181

Michael H. Jones, Ecology, Information Technology, and Environmental Policy: An Ecologist’s Perspective on Values and Value Added, INFORMATION IMPACTS M AGAZINE, Oct. 1999, at http://www.cisp.org/imp/october-99/10-99huntjones.htm (discussing how on-line databases assist in environmental policymaking and identifying numerous on-line environmental data bases, including electronic sites for the Ecological Society of America, the US Global Change Research Program, and the National Snow and Ice Data Center). In important respects, environmental analysis is simply catching up with other social sciences, such as economics and even political science, where data-driven studies have become the norm in recent decades. Daniel C. Esty, Toward Data-Driven Environmentalism: The Environmental Sustainability Index, 31 ENVTL L . REP . 10603 (2001). 182

Howard Latin, Good Science, Bad Regulation and Toxic Risk Assessment, 5 YALE J. REG 89, 90 (1988) (dismissing the possib ility of purely scientific or rational environmental decision making). 183

Mark Sagoff, We Have Met the Enemy, and He is Us or Conflict and Contradiction in Environmental Law, 12 ENVTL. L. 283, 286-97 (1982) (arguing that valuations inevitably turn on value judgments).

59

on which valuations are based can be spelled out, made public, opened to review, and debated—all of which is likely to narrow differences of opinion, at least over time. 5. De-collectivization In some number of cases, information technologies offer the prospect of promoting individual environmental decisionmaking in place of “policymaking” with its susceptibility to error, inefficiency, and special interest capture. Such a shift makes sense where risks and impacts can be internalized at the individual or household level. De-collectivization is particularly useful where tastes and values vary and where individuals are capable of making informed choices in a market context. Under such an approach, individuals decide for themselves how to weigh the competing risk and cost factors. The regulatory role is limited to providing reliable information. Food safety stands out as a realm where some choices that are presently made in a regulatory setting can be devolved to the individual. For instance, the decision whether to buy food derived from genetically modified organisms or to purchase GMO-free products may be better left to individuals rather than regulators.184 Where individual choice represents the preferred environmental protection strategy, ecolabels or other information-provision strategies make sense as a way to promote informed preferences. 6. Contracting Beyond Small Numbers The prospect of appropriate environmental outcomes being negotiated, where harm-causers compensate those whose rights have been infringed (or where pollutees pay polluters to reduce the harms), has long been understood to diminish beyond the realm of small numbers.185 But the Internet and the growing network of other Information Age mechanisms for linking people together may change what counts as “small numbers,” and thus redefine this dimension of the transaction cost calculus.186 Environmental rights holders may soon be able to organize themselves at much lower cost than 184

John Charles Kunich, Mother Frankenstein, Doctor Nature, and the Environmental Law of Genetic Engineering, 74 S. CAL. L. REV. 807 (2001) (explaining why GMO regulation should have an information focus). 185

Coase, Social Cost, supra note 2 (discussing the parameters for successful bargaining).

186

John Hagel III and Marc Singer argue one of the greatest values of the Internet is its capacity to connect people in a relevant and timely fashion. JOHN HAGEL III & M ARC SINGER, NET W ORTH xii (1999).

60

historically, enabling them (or their agents) to negotiate more systematically and effectively. 187 One can imagine, for example, that those seeking to site a new factory may be able to negotiate electronically with the neighbors over compensation for any new harms. There are, however, real limits to how much environmental information individuals can process.188 Some dimensions of environmental problem solving are inescapably technically complex. The presence of good data and analysis does not guarantee good decisions. Expert analysis may therefore be required to frame intelligible choices. How much environmental choice individuals want may also be limited. In a world with endless distractions, many people may choose to spend their free time playing computer games rather than on environmental decisionmaking. Simply put, how many individuals would want to negotiate compensation agreements with the polluting factories in town? Or even find an agent to do so? The combination of human cognitive limitations, an environmental problem set that is technically complex, and a rational choice on the part of many citizens to limit the time they spend on environmental matters circumscribes the ultimate scope of pollution control and natural resource management based on Coasian bargaining. Thus, while we can anticipate a more robust environmental marketplace in the future, a pure “property rights” approach to environmental protection seems unlikely anytime soon. B. Court Protected Entitlements Insofar as the shift toward regulation arose, in part, because of frustration with the inability of courts to overcome information gaps in environmental cases,189 better analytic underpinnings for court187

See e.g., Paul M. Schwartz, Privacy and Democracy in Cyberspace, 52 VAND . L. REV. 1609, 1647-1650 (1999) [hereinafter Schwartz, Cyberspace] (noting that “cyberspace has the potential to emerge as an essential focal point for communal activities”); Jocelyn C. Adkins, Environmental Database Review: The Internet: A Critical Technology for the State of Environmental Law, 8 VILL . ENVTL. L. J. 34 (1997) (describing the broad availability of environmental information). 188

Jolls et al., A Behavioral Approach, supra note 10 (discussing the difficulties individuals have in making accurate risk assessments). 189

In a series of 19th Century pollution cases both in Britain and the United States, courts failed to stop harms emanating from industrial facilities, finding that data to justify intervention had not been brought forward. See, e.g., Hole v. Barlow, 140 Eng. Rep. 1113, 4 C.B. (N.S.) 334, 335 (C.P. 1858); St. Helen’s Smelting Co. v. Tipping, 11 Eng. Rep. 1483, 11 H.L. Cas. 642 (1865); Sparhawk v. Union Passenger Ry. Co., 54 PA. 40 (1867); Francis v. Schoellkopf, 53 N.Y. 152 (1873). In Huckenstein’s Appeal, for example, the Pennsylvania Supreme Court rejected a special master’s conclusion that sulphuric

61

based decisionmaking may reverse this trend. In fact, all Information Age advances outlined in Part IV will facilitate not just a shift on the margins toward a contractual approach to environmental protection but will also narrow the scope of environmental information gaps in ways that facilitate judicial decisionmaking. These potential gains are important because a well functioning tort system is a necessary backstop for a property rights based environmental regime. As discussed above, a more information-rich world eases the burden of case-by-case analysis and thus reduces the need for collectivized decisionmaking. This shift in “optimal specificity” and easy access to vast stores of comparative data also makes it easier to spot outlier valuations, bring pressure to bear on holdouts, and discipline strategic behavior — facilitating adjudication of contract disputes. The tend toward greater information availability will thus reduce the risk of bargaining failures and lower other transaction costs, which may permit greater reliance on property rather than liability rules. More generally, a better functioning system of contracts and tort law makes a property rights approach to environmental protection more viable. C. Command and Control Regulation Improved information will not mean perfect information across the full spectrum of uncertainties any time soon. This reality negates the possibility of an environmental protection system based entirely (or even primarily) on market exchanges of property rights. But enhanced environmental information promises to make it possible to shift to a more empirical and refined approach to environmental protection. Thus, to the extent that command and control regulation persists, we can antic ipate better results with regard to each of the five primary categories of regulatory failure identified in Part III — (1) technical deficiencies; (2) cognitive shortcomings; (3) structural (or jurisdictional) mismatches; (4) public choice failures; and (5) administrative inefficiencies.

fumes from the defendants brickyard caused plaintiff’s harms, suggesting that a sufficient causal link had not been established. Huckenstein’s Appeal, 70 PA. 102, 107 (argued 1871, decided 1872). The broader case history of this period is replete with decisions turning on problems of harm identification, causation, valuation, and other elements of the information gap taxonomy developed above. For a complete discussion, see Joel Franklin Brenner, Nuisance Law and the Industrial Revolution, 3 J. LEGAL STUD. 403 (1974); see also Rosen, Differing Perceptions, supra note 127.

62

1. Technical Deficiencies Regulatory mistakes that can be traced to incomplete or inaccurate data and analysis should diminish steadily. The same Information Age breakthroughs that make markets work better will also make it easier for regulators to spot issues, identify problems, analyze the fate and transport of chemicals, measure environmental harms, undertake risk assessments, carry out cost-benefit analyses, understand interactive effects including both cumulative pollution exposure impacts and problem-reducing synergies, track the time lags before effects emerge, sharpen evaluation techniques, deal with diversity, and manage complexity. Simply put, regulatory efficacy is poised for dramatic improvements as digital technologies are embedded in the spectrum of activities that make up the policymaking process. Fundamentally, digital technologies allow us to “see” more. With computer chips (controlling sensors, monitors, and electronic reporting devices) embedded in everything from smokestacks to automobile tailpipes the basic task of figuring out who is emitting what promises to become easier and vastly cheaper. We will, for example, be able to track volatile organic compounds (VOCs) on a car-by-car basis not just as an aggregate level across an airshed. Just as sensors make it easier to see emissions, new statistical techniques (e.g., advanced regression methodologies and neural net analysis) 190 make it easier to spot patterns of pollution harm and establish causal connections.191 These statistical advances allow investigators to trace multiple causes as well as understand cumulative impacts and synergies. We are, for instance, learning how timbering practices, such as clear cutting, affect levels of sunlight on wetlands, which in turn affect habitat

190

Recent computational advances have vastly increased the power of many statistical techniques. See, for example, the websites of the Neural Network Society, at http://ieee-nns.org/general/, and the International Neural Network Society Journal, at http://www.inns.org/nn.html, for a sampling of the literature on the latest computer-aided advances in neural networking and its widely expanded commercial applications. See also , Statistical Primer, U.S. Environmental Protection Agency, at http://www.epa.gov/bioindicators/primer/index.html (detailing the EPA’s efforts to apply regression analysis and other recent advances in statistical modeling techniques to aquatic resources and biodiversity management). 191

Such modeling has transformed the debate over the sources of air pollution in the Northeastern United States. For an example of ozone modeling, see the final report and technical supporting document produced by the Ozone Transport Assessment Group, at http://www.epa.gov/ttn/otag/finalrpt (last visited Apr. 5, 2003).

63

productivity and species diversity. 192 Thus, while the complexity of the environmental realm will not diminish, our ability to make sense of all that is going on looks likely to vastly increase.193 Valuation methodologies are undergoing a similar computer-aided revolution. New information management technologies permit much greater quantities of data to be assembled and analyzed, which makes it possible to do cost-benefit analysis on a more individualized and sophisticated basis.194 As environmental decisionmaking becomes more empirical, the likely trajectory of ecological and epidemiological impacts and costs becomes easier to plot. As noted above, a more data-driven policymaking process will not eliminate valuation disputes, but Internetfacilitated data access will make it much easier to find comparable cases as a benchmark. The shift toward a more empirical approach to environmental decisionmaking entails not only more data on pollution problems and trends but also better tools for performance measurement that will allow decisionmakers to gauge policy success. Comparative data positions policymakers to identify and replicate successful intervention strategies and to abandon those that do not deliver good results. 195 As performance data becomes more refined and accessible, decisionmakers will be positioned to hone in ever more precisely on comparable circumstances. A homeowner trying to figure out, for example, how well his family is doing on energy conservation will soon be able to get data not just on last month’s or last year’s energy use at his own house but on the energy efficiency of other households of comparable size in

192

M.A. Halverson, et al., Forest Mediated Light Regime Linked to Amphibian Distribution and Abundance, 134 OECOLOGIA 360-64. 193

We will be better positioned to understand and address anomalous circumstances and optimize “slippage” from general rules. Daniel A. Farber, Taking Slippage Seriously: Noncompliance and Creative Compliance in Environmental Law, 23 HARV. ENVT ’L. L. REV . 297 (1999); see also Dennis D. Hirsch, Project XL and the Special Case: EPA’s Untold Success Story, 26 COLUM. J. ENVT ’L. L. 219 (2001) (informational advances may reduce the need for waivers from general rules). 194

See, e.g ., Raymond J. Kopp, et al., Cost-Benefit Analysis and Regulatory Reform: An Assessment of the Science and the Art, RESOURCES FOR THE FUTURE (Jan. 1997), available at www.rff.org/disc_papers/PDF_files/9719.pdf (examining the “state of the science” with respect to cost-benefit analysis, with special emphasis on its application in the environmental domain). 195

EPA recognizes this potential and has begun to invest heavily in data and metrics. See the range of information resources at www.epa.gov (last visited May 6, 2003).

64

a similar geographic zone.196 Benchmarking against peers highlights not just one’s own relative performance but also what constitutes “best practices.” As noted earlier, such analysis offers powerful motivation to those whose performance is lagging — and provides critical information on intervention options, implementation, and policy refinement.197 2. Cognitive Shortcomings In a number of areas, environmental progress has been stymied by the difficulty of motivating action in the face of disputed views about a problem. Part of the issue in such cases may arise from the limits of the human mind to process risk information, take unseen problems seriously, understand points of environmental leverage, and agree on appropriate action in response. Many people, for example, have trouble appreciating the threat of climate change with its geographically and temporally diffuse impacts on global temperatures, rainfall patterns, sea levels, and storm intensities. Perception problems can be difficult to overcome, but they may be mitigated with better data and sophisticated information displays. 198 For example, public understanding of the problem presented by the thinning of the ozone layer greatly improved when vivid maps showing the Antarctic ozone “hole” were produced. As noted above, Information Age breakthroughs offer a powerful tool for sharpening our environmental focus and overcoming cognitive failures. A more empirical and data-driven approach to environmental protection offers a promising avenue for overcoming the exaggerated emphasis given to recent, sensational, or problems that resonate as well as those brought to prominence by availability cascades triggered by the media or those with special interests to advance. Gains in “realization” will come not only from direct advances in the ability of decisionmakers to trade emissions and see harms, but

196

First steps toward a broader world of comparative performance measurement are already being taken. In Finland, for example, electric bills come with data not only on the prior month’s and last year’s consumption, but also with benchmarks on how one’s energy use compares to others. (Sample bills on file with the author). 197

Daniel C. Esty, Why Measurement Matters, in ENVIRONMENTAL PERFORMANCE M EASUREMENT : THE GLOBAL 20012002 REPORT (Daniel C. Esty and Peter Cornelius, eds. 2002), at 5-6. 198

EDWARD R. TUFTE, THE VISUAL DISPLAY OF QUANTITATIVE INFORMATION (1983) (demonstrating how sophisticated data displays can change perception).

65

also from improved statistical techniques that will ease the focus on the “here and now,” reducing the impact of the availability heuristic. Likewise, a more quantitative approach to pollution control and natural resource management may bring to light issues that would otherwise remain hidden and ignored. No one could see the ozone layer breaking down as chlorofluorocarbons enter the stratosphere. Computer-supported sensors were, however, able to measure the thinning of the Earth’s protective ozone shield. More generally, data have the capacity to unveil hidden problems, make the invisible visible , and narrow the zone of policy uncertainty. 3. Structural Mismatches Environmental problems exist at an enormous diversity of geographic scales.199 To the extent that the “matching principle”200 —a core element of public goods theory and therefore optimal environmental governance— argues for regulatory intervention at the same scale as the problem, we face the prospect of a nearly infinite number of “optimal environmental areas.”201 While the administrative costs of having regulating authorities available at every such scale would be exorbitant, the relative burden of a multi-tier environmental governance structure goes down as information costs fall, making a regulatory architecture that more closely matches the scale of problems more affordable. Ease of access to critical technical information and cutting edge policy guidance will increasingly make it possible for local and state officials to handle pollution control and natural resource management challenges successfully. Online training and capacity building will further strengthen the claim for

199

See PRESIDENT ’S COUNCIL ON SUSTAINABLE DEVELOPMENT , SUSTAINABLE AMERICA: A NEW CONSENSUS FOR PROSPERITY, OPPORTUNITY, AND A HEALTHY ENVIRONMENT FOR THE FUTURE vi, 11-23, 83-107 (1996), available at http://clinton4.nara.gov/PCSD/Publications/TF_Reports/amer-top.html (discussing the need to address environmental problems at community, national, and international levels). A discussion of the PCSD’s attention to a range of geographic scales is found in J.B. Ruhl, Sustainable Development: A Five-Dimensional Algorithm for Environmental Law, 18 STAN. ENVTL. L. J. 31, 41 (1999). 200

M ANCUR OLSON, THE LOGIC OF COLLECTIVE A CTION: PUBLIC GOODS AND THE THEORY OF GROUPS (1965) at 48-57; Olson, Strategic Theory, supra note 103, at 479-82; Butler & Macey, Matching Principle, supra note 104. 201

A NDRÉ DUA & DANIEL C. ESTY, SUSTAINING THE A SIA PACIFIC M IRACLE: ENVIRONMENTAL PROTECTION AND ECONOMIC INTEGRATION 123-131 (1997) (developing the concept of “optimal environmental areas”).

66

devolution of environmental responsibilities, which permits the fine-tuning of policy interventions to meet local conditions. Differentiated standards and policies matched to specific circumstances should translate into a more granular regulatory structure and therefore fewer and smaller errors of both underand over-regulation. The ease with which policy complexity can be managed in the Information Age also makes “hybrid” regulatory strategies, which split responsibilities across two or more levels of government, more feasible. Under such an approach, the federal government can be charged with undertaking some aspects of policymaking (e.g., problem identification, dose-response evaluation) while city or state officials take on other aspects (e.g., evaluating intervention options and costs in the light of local circumstances). Once again, lower information costs permit a more complex and carefully tailored policy response that responds to the diversity of environmental problems, settings, and values. 4. Public Choice Distortions Ultimately, technical advances in environmental analysis cannot succeed without parallel political and institutional reforms. The capacity for more refined internalization of externalities must be matched with the political will to do so. Special interest lobbying must be reined in. “Captured” regulators must be disciplined. Self-serving bureaucrats must be unmasked. While no silver bullet, the information flow and connectivity of the Digital Age may help to address these challenges. A world of greater transparency and easy cross-jurisdiction policy comparisons tends to highlight subpar results, analytic errors, and distortions in regulatory decisionmaking. Digital technologies may also provide alternative pathways for political dialogue and new ways to “roll up” public preferences and achieve collective action. 202 Historically, the information of collective judgments was almost always mediated by politicians and government structures.203 In the Digital Age, 202

See David Johnson, The New Civic Virtue of the Internet (2002) (available at www.cli.org ) (discussing new governance models). 203

Ostrom has demonstrated that there are a variety of institutional arrangements that can be constructed to achieve collective action. ELINOR OSTROM, GOVERNING THE COMMONS : THE EVOLUTION OF INSTITUTIONS FOR COLLECTIVE A CTION (1990). But, historically, politics and government have almost always played some role in the process.

67

this pattern may not hold. New processes of social ordering may emerge over the Internet whereby individuals and groups converge on-line, engage in information exchange and debate, and evolve toward common goals and strategies. Government action may follow rather than lead this process. The movement to ban landmines, driven largely by a network of NGOs, demonstrates the potential of this new mode of collective action. 204 The Internet and computer power also allows for more complicated processes for mapping out preferences and establishing public priorities.205 The capacity to track numerous data points simultaneously provides a basis for more sophisticated scoring systems and voting techniques that will facilitate decisionmaking across seemingly incommensurate issues and values and permit account to be taken of the intensity of preferences. More data and vivid analysis may help both individuals and interest groups to “see” more and to spot opportunities for agreement. Computer-aided deliberative processes and digitally-enabled graphical display systems may also bring to light previously hard-to-see clusters of overlapping interests and help to spotlight “outlier” positions.206 These tools offer the promise of new ways to achieve consensus, expose self-serving actions, eliminate strategic behavior, and reduce bargaining breakdowns. A more information-rich regulatory regime cannot put a stop to special interest manipulation of the policymaking process or bad political judgments. But greater transparency can expose self interest, more quic kly reveal the implications of the choices being made, and help to reinforce pressures for better 204

The nongovernmental International Campaign to Ban Landmines is an Internet-supported network of a thousand or more NGO affiliates throughout the globe who lead the campaign for the passage, implementation, and monitoring of the 1997 land mine treaty. For more information on this nongovernmental mode of collective action, see the website of the ICBL at http://www.icbl.org/ and the official 1997 Nobel Peace Prize press release at http://www. Nobel.se/peace/laureates/1997/press.html. 205

HOWARD RHEINGOLD, SMART M OBS: THE NEXT SOCIAL REVOLUTION (2002) (discussing the transformation of communities and social activism “in the age of instant access”); Clay Shirky, Social Software and the Politics of Groups (2003) (available at www.shirky.com last visited May 6, 2003) (discussing how the Web refines group dynamics and human interaction); John Clippinger & David Bollier, A Renaissance of the Commons: How the New Sciences and Internet are Framing a New Global Identity and Order (2003 unpublished manuscript on file with author) (discussing how the Internet is changing group identity and decisionmaking). 206

See JAMES S. FISHKIN, THE VOICE OF THE PEOPLE (1995) (discussing “deliberative polling”). See also Beth Simeone Noveck, Designing Deliberative Democracy in Cyberspace: The Role of the Cyber-Lawyer, 9 B.U. J. SCI. & TECH. L. 1 (2003) (contemplating the possibility of conducting new ways to survey citizen opinion in cyberspace).

68

performance. In circumstances where comparative data is hard to find, anomalous policy judgments may go unchallenged for long periods. Where information is plentiful, policy choices that deviate from established best practices are likely to stand out much more quickly. For example, Belgium has suffered from poor environmental conditions for decades. But, when the 2001 World Economic Forum Environmental Sustainability Index placed the country in 79th rank, just behind Albania, an uproar ensued in Brussels. Environmental groups, the media, and opposition politicians all pressed the government to explain the nation’s relatively poor pollution control performance.207 In effect, cheaper and better data makes the intellectual marketplace undergirding the regulatory process operate more efficiently. In a similar fashion, easy access to information on who is participating in the decisionmaking process can throw a spotlight on special- interest manipulation of the political process.208 With readily available reports on who is lobbying, it becomes much easier to identify and draw attention to instances of regulatory capture or other cases where disproportionate influence is being wielded by certain groups or individuals. Similarly, in jurisdictions where lavish campaign contributions or even outright corruption taint decisionmaking, Information Age tools (e.g., online access to campaign finance reports or officeholder financial disclosure statements) can facilitate disclosure of improper influence, making outcomes that reflect public choice failures harder to sustain. Administrative law could be the legal arena most transformed by Information Age tools. Not only do disclosure statements make it easier to spot improper interest group influence, but the Internet opens up regulatory proceedings. Today anyone with a computer can track critical policy debates as they unfold in the public docket. Anyone with email can participate in the Notice and Comment process required by the Administrative Procedure Act. Distance from the physical locus of policymaking becomes much less important.

207

Political inquiries, press reports, and website “hits” are on file with CIESIN, one of the groups involved in the ESI rankings. See www.ciesin.columbia.edu. 208

The Common Cause website, for example, allows users to access campaign contribution profiles of individual politicians, as well as search a database of special interest soft money contributions to the major political parties. See www.commoncause.org (last visited June 1, 2001).

69

Expanded participation could significantly democratize the regulatory process and give people with good ideas — even those who never get to Washington (or wherever the regulators are located) — a chance to shape policy outcomes.209 This trend bodes well for the environment as there is a growing scholarly literature that connects good environmental results with the strength of a jurisdiction’s democratic institutions and the robustness of public debate.210 5. Administrative Inefficiency Cheaper and better information also makes regulatory efficiency easier to achieve. As the cost of precision falls, uniform rules and standards based on gross-average data can be replaced with ones that vary with individual circumstances. Shortcuts taken to reduce administrative costs become less necessary. Both over- and under-regulation can thus be reduced. The Digital Era’s powerful dissemination tools — Internet, email, list servers — makes it possible for both regulators and the regulated community to more easily identify and adopt “best practices,” including policies, technologies, and regulatory approaches from distant places across the country and around the world. A proliferation of environmental indicators, measurement initiatives, and data sets makes it easier to benchmark governmental performance,211 facilitating “regulatory competition” and reducing regulatory inefficiency. 212

209

Whether Information Age decision processes will be more representative of underlying public opinion remains uncertain. Easy access to the political process may result in “loud” interests having an easier time wielding influence. But the mechanisms for screening inputs into the decisionmaking processes and discounting those from special interests may also get better. 210

Scott Barrett & Kathryn Graddy, Freedom, Growth and the Environment, 5 ENVT . & DEV. ECON. 433-456 (2000) (demonstrating the impact of civil and political freedoms on environmental quality); Esty & Porter, Measuring, supra note 211 (showing empirical correlation between legal and political institutions and environmental results). 211

Initial efforts to rank governments based on their environmental performance have begun to emerge. See World Economic Forum, Pilot Environmental Sustainability Index (2000); Daniel C. Esty & Michael E. Porter, Measuring National Environmental Performance and Its Determinants, in GLOBAL COMPETITIVENESS REPORT 2000 60-75 (Michael Porter et al. eds., 2000) [hereinafter Esty & Porter, Measuring]. 212

For a review of regulatory competition theory, see Richard Revesz, Federalism and Regulation: Some Generalizations, in REGULATORY COMPETITION, supra note 21; see also Charles M. Tiebout, A Pure Theory of Legal Expenditures, 64 J. POL. ECON. 416 (1956).

70

The power of comparative data is especially important to note. Quantitative performance measures drive companies, communities, and even individuals to gauge their relative environmental performance against relevant peer groups. Executives in corporations where emissions per unit of output lag industry averages find themselves under scrutiny and pressed to do better by NGOs, the media, nearby communities, and their own employees, not to mention regulators. Likewise, mayors seek to move their communities to the top of the “Recycling League Table.” And most individuals want to be sure that their cars, houses, or personal consumption habits do not set them apart as environmental miscreants. Wider flows of information increase the number of intellectual competitors in the environmental policymaking marketplace and multiply the vectors of regulatory competition. The pressure on governments to achieve better environmental performance increasingly comes not just from Tieboutian “horizontal” interactions (e.g., California’s innovations being copied by Oregon, cutting-edge U.S. strategies being emulated by the European Union), but also from “vertical” competition (and cooperation) between levels of government. With easier access to comparative data, policy successes at the state level or international level can stimulate “copycat” federal legislation or regulations. 213 Likewise, superior regulatory performance at a local or federal scale keeps state environmental decisionmakers on their toes. Greater access to information and thus a broadened capacity for leading-edge analytic work creates the potential for a welfare-enhancing mix of competition and cooperation between governments and non-governmental organizations, including environmental groups and the business community. 214 Although there is some risk of cacophony or information overload (discussed in Part VII below), input from alternative perspectives generally facilitates problem solving. Multiple sources of data and analysis

213

The 2002 adoption in California of a law mandating reduced greenhouse gas emissions from vehicles, in the face of federal government inaction on climate change, provides a vivid example of this dynamic. See A.B. 1493, 2001-2002 session (Ca. 2002) (adding § 43018.5 to the California Health & Safety Code, approved by Governor July 22, 2002). 214

M ARK BLAUG, THE ECONOMIES OF EDUCATION AND THE EDUCATION OF AN ECONOMIST 41-2 (1987) (stressing that the essence of science and modern knowledge development lies in intellectual competition); Daniel C. Esty & Damien Geradin, Regulatory Co-opetition, 3 J. INT ’L. ECON. L. 235 (2000) [hereinafter Esty & Geradin, Co-opetition] (discussing the value of both competition and cooperation).

71

are especially valuable under conditions of uncertainty as they permit decisionmakers to triangulate on reality and to get a better fix on their policy options. D. Market-Based Regulation Economic -incentive-based regulation represents, in effect, a way-point between the informationintensity required for a fully functioning property rights regime and the broad-scale collectivized decisionmaking of command and control regulation. The same Information Age breakthroughs that create the potential for a more vibrant market in environmental rights (discussed in Part V. A. above) provide the promise of more immediate gains through expanded reliance on market-based regulatory tools. Information technologies may help to address a number of the obstacles that have prevented broader adoption of market-based regulatory strategies. 215 Information technologies can ameliorate valuation disputes, reduce search costs, make property rights easier to define and defend, and generally lower transaction costs. Most notably, sensors and communications tools make measuring emissions and tracing harms much easier. Quantitative measurement, in turn, facilitates the internalization of externalities using pricing mechanisms. In the last decade, market forces have been harnessed to environmental objectives using both Pigouvian pollution fees and emissions allowance trading schemes in a number of critical issue arenas from control of ozone-layer-damaging CFCs to acid rain. 216 To the extent that complexity has been a major obstacle to more widespread use of market-based regulation, advanced information systems, with their ability to track multiple variables simultaneously, can be at least a partial answer. Wetlands “banking” is made much easier, for example, with sophisticated databases that track, price, and make available for purchase of various wetlands restoration and creation projects so that

215

Nathaniel Keohane et al., The Positive Political Economy of Instrument Choice in Environmental Policy, in ECONOMICS ENVIRONMENT : SELECTED READINGS (2000).

OF THE 216

J. Andrew Hoerner, Taxing Pollution, in OZONE PROTECTION IN THE UNITED STATES : ELEMENTS OF SUCCESS (Elizabeth Cook ed., 1996); Clean Air Act Amendments §403, 42 U.S.C. 7651(b) (1990) (authorizing acid rain allowance trading).

72

developers can more easily and precisely offset damage they might be causing. As all these technologies improve, systems for emissions tracking, charging, and trading should be available more broadly. 217 Although political obstacles undoubtedly stand in the way, we are not very far away, as a technical matter, from being able to establish a marginal cost for any and all emissions harms. This capacity creates enormous opportunities for new approaches to environmental protection. Laser beams which can detect and analyze vehicle exhaust from a distance are now coming on to the market.218 One can therefore imagine monthly car pollution bills being sent out alongside utility charges. The leading edge of this trend can already be seen in places like Singapore, where commuters pay electronicallydeducted tolls with differential charges, depending on how far into the urban center they take their cars and at what time of day. 219 At some point, multiple -variable fees might become confusing to the paying public, but the chance to tailor incentives with considerable precision represents a major policy opportunity. Certain market approaches to environmental protection, notably emissions allowance trading regimes, have traditionally been seen as vulnerable to the possibility that either the wrong number of permits would be issued to achieve the desired level of control or that the price of the permits would fluctuate wildly. 220 Information technologies can be used to generate valuable supply and demand 217

In the closing days of his Administration, President Clinton called for CO2 and mercury emissions controls through new tradeable permit regimes. See GREENWIRE, Nov. 14, 2000. The Bush Administration similarly wants to expand the use of market-based mechanisms. See, for example, the “Clear Skies” proposal. Darren Samuelsohn, Bush’s Clear Skies Plan Reintroduced in Both Chambers, ENVIRONMENT & ENERGY DAILY, Feb. 28, 2003. 218

See, e.g., M. A. Gondal, Laser Photoacoustic Spectrometer for Remote Monitoring of Atmospheric Pollutants, 36 A PPLIED OPTICS 3195 (May 1997); see also Ion-Optics website (2000), at www.ion-optics.com (highlighting emerging laser beam based technologies). 219

Felicia B. Young & John T. Berg, Value Pricing Helps Reduce Congestion, 62 PUB . ROAD 47 (March 1, 1999). The Singapore experience suggests that pollution charges can be set to reflect not only the precise amount of pollution coming out of the tailpipe of a particular vehicle but also varied according to the time of day as well as the geographic locale in which one is driving — both factors which affect the level of harm inflicted and thus the appropriate charge under an economically-optimal marginal cost pricing scheme. London has recently adopted a similar vehicle pricing regime for the center city. See, generally, Randy Kennedy, The Day the Traffic Disappeared, NEW YORK TIMES M AGAZINE (April 20, 2003). 220

See TOM TIETENBERG, ENVIRONMENTAL ECONOMICS & POLICY 278 (2001); B.S. Fisher et. al., An Economic Assessment of Policy Instruments for Combating Climate Change, in CLIMATE CHANGE 1995: ECONOMIC AND SOCIAL DIMENSIONS OF CLIMATE CHANGE 397, 403 (J. P. Bruce et al., eds., 1996) (discussing the possibility of implementing a tradable-permit system in combination with an international carbon tax, and the potential for under- or over-abatement).

73

information, speed up feedback loops, and thus facilitate quicker “clearing” of emissions markets. Similarly, when environmental taxes are employed as a pollution control measure, there is no guarantee that they will be set at the level required to achieve the desired degree of emissions control. Again, digital technologies provide a capacity to monitor market behavior on a real time basis, including behavioral changes induced by economic incentives. By tracking activity patterns, regulators will be able to quickly adjust tax rates to achieve desired emissions levels. Fundamentally, cheaper, faster, better, and more accessible information makes markets work more efficiently — and concomitantly enhances the utility of market-based regulatory tools. E. Social Context and New Modes of Collective Action Broad-based community norms can shape behavior even in the absence of legal structures. 221 The environmental movement of the past 30 years has done a great deal to change the social context within which individual and business choices are made. Although the emergence of environmental protection as a societal goal predates the Information Age, various digital advances seem likely to reinforce the inculcation of environmental values and may help to speed up the norm development process. Corporations, in particular, face a new era of transparency in which their environmental performance is open to scrutiny by “green” consumers, capital market analysts (some of who see a correlation between environmental performance and financial results),222 corporate sustainability rating

221

Robert Ellickson has chronicled a number of circumstances where a formal legal structure is not n ecessary and externalities are internalized on the basis of community norms. See Robert C. Ellickson, Law and Economics Discovers Social Norms, 27 J. LEG. STUD. 537 (1998) [hereinafter Ellickson, Social Norms] (“informal systems of external social control are far more important than law in many contexts”). ROBERT C. ELLICKSON, ORDER W ITHOUT LAW : HOW NEIGHBORS SETTLE DISPUTES (1991) (norms may be widespread and powerful even without the force of law); see also Eric L. Talley, Disclosure Norms, 149 PENN. L. REV. 1955 (2001) (arguing that extra-legal norms and legal rules are more than policy substitutes and that norms and laws can help to reinforce each other); FLORINI, COMING DEMOCRACY, supra note 34 (discussing how transparency and disclosure strengthen norms). 222

G. Dowell et al., Do Corporate Global Environmental Standards Create or Destroy Market Value?, 46 M GMT . SCI. 1059 (2000) (finding that positive market valuation is associated with the adoption of single stringent environmental standards); Shameek Konar & Mark A. Cohen, Does the Market Value Environmental Performance?, 83 REV. ECON. & STAT . 281 (2001) (finding that reduced emissions of toxic chemicals are correlated with higher market value); Michael V. Russo & Paul A. Fouts, A Resource-Based Perspective on Corporate Environmental Performance and Profitability, 40 A CAD. M GMT . J. 534 (1997) (reporting a correlation between environmental performance and economic performance, stronger in higher-growth industries).

74

agencies,223 community activists,224 NGOs,225 industry associations,226 and environmentally conscious employees. Some of these groups may even become “norm entrepreneurs,” consciously working new avenues of information dissemination to re-shape public values.227 As these “social context” pressures mount, many companies have developed sophisticated environmental programs to push their performance beyond compliance with regulatory requirements and to meet the high expectations of their customers, employees, and other stakeholders.228 Aided by the emergence of information-intensive environmental management systems, including ISO 14000,229 the European Union’s Eco-Management and Auditing Scheme (EMAS),230 and the Global Reporting Initiative (GRI),231 a growing number of companies are making environmental protection or sustainability

223

Alois Flatz, Corporate Sustainability and Financial Indices, in ENVIRONMENTAL PERFORMANCE M EASUREMENT , supra note 110, at 66-81; Frank Dixon, Financial Markets and Corporate Environmental Results, in ENVIRONMENTAL PERFORMANCE M EASUREMENT , supra note 110, at 54-65. 224

These activists are empowered by easy-access TRI data and other information such as the Environmental Defense “Scorecard.” See Frances Irwin and Carl Bruch, Information, Public Participation, and Justice, 32 ENVTL L. REP . 10784 (2002). 225

The Non-governmental Order: Will NGOs Democratize or Disrupt Global Governance?, ECONOMIST, Dec. 11, 1999, at 20 (discussing NGOs’ increasing political power and use of the Internet). 226

Daniel J. Fiorino, Rethinking Environmental Regulation: Perspectives on Law and Governance, 23 HARV. ENVTL. L. REV. 441, 454-55 (noting that programs like the Chemical Manufacturing Association Responsible Care “constitute efforts to improve the behavior of firms in different sector and commit associations and their members to a more public responsibility for the environment.”) For an overview of Responsible Care principles and requirements, visit the Responsible Care Practitioners Site at www.cmahq.com/rc.nsf/open?OpenForm (last visited June 13, 2001). See also International Council of Mining and Metals “Sustainable Development Charter,” available at http://www/icmm.com/html/charter_intro.php. 227

Cass R. Sunstein, On the Expressive Function of Law, 144 U. PA. L. REV. 2021 (1996) (discussing the role of “norm entrepreneurs”); see also Dan Kahan, The Secret Ambition of Deterrence, 113 HARV. L. REV. 413 (1999) (describing how “norm entrepreneurs” and “expressive zealots” work to inculcate social norms and mold public values). 228

Leslie Carothers, Integrating EMS into Manufacturing, ENVTL F. (1999); BRUCE SMART , BEYOND COMPLIANCE: A NEW INDUSTRY VIEW OF THE ENVIRONMENT , World Resources Institute (1992) (explaining the logic of voluntary environmental efforts). 229

ISO 14000 AND BEYOND (C. Sheldon ed., 1997) [hereinafter ISO 14000]; see also M ARC EPSTEIN, M EASURING CORPORATE ENVIRONMENTAL PERFORMANCE (1996). 230

A. Spencer-Cooke, From EMAS to SMAS: Charting the Course from Environmental Management to Sustainability, in ISO 14000 AND BEYOND, supra note 229. 231

See www.ceres.org and www.globalreporting.com.

75

part of their mission.232 Recent corporate scandals and the growing momentum for more “corporate responsibility” — often driven by Internet-based pressure groups 233 — is likely to add to the values-based pressures for environmental care. The Internet furthermore fuels “information arbitrage” whereby multinational corporations face questions everywhere about their actions anywhere. A company with a poor environmental record in a distant corner of its worldwide empire will find its reputation suffering globally not just where the harm has occurred. For example, Rio Tinto, an Anglo-Australian mining concern, has come under attack in Europe and the United States for exposing workers to radiation at its uranium mine in Namibia and for causing cyanide to leach into nearby waters at its goldmine in Kalimantan, Indonesia.234 F. Institutional Realignment While Digital Age advances offer the promise of improving the efficacy of all of our existing strategies for addressing environmental harms by ameliorating information gaps, a shift in emphasis toward some institutional mechanisms and away from others seem likely. Lowered transaction costs change the “optimal specificity” tradeoffs decisionmakers face and thus favor certain regulatory strategies. 1. Decollectization and the Rise of Market Approaches As a general matter, better and cheaper data reduces the risk of information-based market failures. Lower cost information changes the optimal specificity of environmental decisionmaking. Precision becomes cheaper, and the need to handle issues at a broad scale and to rely on gross averages in analysis and decisionmaking is therefore reduced. This shift makes reliance on contractual approaches to

232

FOREST REINHARDT , DOWN TO EARTH (2001) [hereinafter REINHARDT , DOWN TO EARTH] (discussing corporate environmental management pressures); GREEN LEDGERS: CASE STUDIES IN CORPORATE ENVIRONMENTAL ACCOUNTING (D. Ditz et al. eds., 1995); Andrew J. Hoffman, Trends in Corporate Environmentalism: The Chemical and Petroleum Industries, 1960-1993, 9 SOC ’Y & NAT . RESOURCES 1, 47-64 (1996); Andrew J. Hoffman, Institutional Evolution and Change: Environmentalism and the U.S. Chemical Industry, 42 A CAD. M GMT . J. 351 (1999). 233

See, e.g., www.corpwatch.org, www.stakeholderdemocracy.org, www.globalwitness.org, www.hrw.org, and www.foe.org. 234

See www.corpwatch.org/compaigns/PCD.jsp?articleid=622 (last visited April 14, 2003).

76

environmental protection and the use of market mechanisms more attractive. Interventions that directly internalize externalities and protect property rights likewise become possible. These approaches may also be preferable as a normative matter, making it easier to make pollutees whole. While we should not anticipate a wholesale shift to Coasean bargaining over environmental property rights, some marginal movement in this direction is likely. New technologies, as discussed above, will make it possible to partition some collective resources, from the oceans (fisheries) to the atmosphere (greenhouse gas emissions), allowing for more efficient natural resource management. Within the pollution control sphere, information technologies will encourage a parallel shift toward market-based regulation. Similarly, reliance on “command and covenant” approaches to regulation might expand as part of the evolution away from the hierarchy of governmental decisionmaking toward more individualized and tailored regulatory strategies.235 While a shift away from the blunt instrument of command and control regulation will generally increase efficiency, individual choice, and policy efficacy, such a move may not be optimal in all cases. Consider the situation where improved data and analysis highlights a presumptively “correct” answer to an environmental question, but other obstacles (e.g., political intervention by rent seeking interests) obstruct progress. In such circumstances, a governmental mandate may be justified. For example, the environmental superiority of fluorescent lighting (i.e., using less energy and therefore generating less pollution) has now been definitively established, 236 yet inertia and the self-interested posturing of incandescent bulb makers seems to have slowed the shift toward this more efficient new lighting mode. To speed up the process, governmental “commands” such as building codes and procurement guidelines favoring fluorescent lights might be warranted. 235

This approach has long been popular in Europe. Eric Orts & Kurt Deketelaere, Environmetnal Contracts: Comparative Approaches to Regulatory Innovation in the United States and Europe (2001) (highlighting how the covenant strategy is used in the Netherlands and elsewhere in Europe). The trend toward environmental covenants can now be seen in U.S. legislation. The Second Generation of Environmental Improvement Act H.R. 3448 (1999) promoted both information generation (Title I) and authorization for individualized compliance agreements (Title II) a la the Clinton Administration’s Project XL. 236

See EPA’s Green Lights program, at http://www.epa.gov/greenlights. See also Arnold W. Reitze Jr., Global Warmi ng, 31 ENVTL L. REP . 10253 (2001). Of course, in some cases, the “quality” of the light may matter to the consumer.

77

2. The End of “Do Nothing” The most dramatic shift in the current pattern of environmental responses will be a significant reduction in the number of cases where “doing nothing” is deemed optimal. As lower-cost and easieraccess information results in a re-calibrated optimal specificity of regulation, a series of practices and policy presumptions, which have been deployed to rationalize a “do nothing” environmental strategy, will have to be rethought. a. No Visible Harm Means No Harm It is perhaps human nature to want to see, smell, hear, taste, or feel the purported link from harm to victim as a test of whether the problem is real and as a way to judge its magnitude. But the dismissal of harms that cannot be traced by the human senses is intellectually and ecologically curious if not indefensible.237 Modern science permits us to detect many things that an average person cannot sense.238 Nevertheless, the maxim “out of sight, out of mind” continues to hold significant sway in the environmental realm. 239 More troublingly, the out-of-sight-means-no-big-harm legend frequently means limited policy attention. Take indoor air pollution. No one sees the problem, but the harms from invisible gases and chemicals are real and increasingly well documented. 240 The risks are hard to perceive, and the public is not clamoring for action. Thus, policymakers have tended to skip over the

237

DAWES , IRRATIONALITY, supra note 132 (explaining how and why people make such errors).

238

Technologies, which detect unseen pollutants are being developed by universities, governments, and private companies. For instance, see the Heterodyned Holographic FTS Project at Imperial College, available at http://op.ph.ic.ac.uk/users/tony/hhfts.html (proposing a portable device that measures ultra-violet light to detect nitrogen dioxide, sulfur dioxide and ozone). See also U.S. EPA, Regulatory Reinvention (XL) Pilot Projects; Project XL Proposed Project Final Agreement: Progressive Insurance Project Pay-as-you-Drive Auto Insurance, 65 Fed. Reg. 39614 (June 27, 2000) (describing efforts to fit automobiles with pollution sensors). See John Medearis, Skies Finally Clearing for Pollution Control Firm, L.A. TIMES , Oct. 24, 1989, at 9A (describing firm that manufactures stack monitors). 239

GRAHAM & W IENER, RISK V. RISK, supra note 19; Cass R. Sunstein, Economics & Real People, 3 GREEN BAG 2d 397, 400 (2000) (describing the “availability heuristic” and the tendency of the public to assign greater seriousness to those risks more easily called to mind or made visible through publicity); Cass Sunstein, Cognition and Cost-Benefit Analysis, 29 J. LEGAL STUD. 1059, 1067-69 (describing the tendency to overregulate “on-screen” dangers, while ignoring “offscreen” dangers and benefits). 240

EPA, REDUCING RISK: SETTING PRIORITIES AND STRATEGIES FOR ENVIRONMENTAL PROTECTION (1990) (noting the dangers of indoor air pollution).

78

problem and assume that nothing needs to be done. A more data -driven approach to environmental problems is likely to change this pattern. b. Myth of Endless Abundance. If a resource is not scarce or otherwise under pressure, a regime of no controls may well be optimal. 241 In the early nineteenth century, when land seemed endlessly available in the American West, it made sense to permit “free” exploitation of the available resources. The obverse of this principle is that resources under pressure must somehow be allocated or managed to avoid over-exploitation and market failure as well as infringements on property rights. As lower cost, information-based resource management regimes become available, the point at which some system of control system makes sense will emerge earlier (see Figure 1). Figure 1242 Intervention Costs

Do Nothing Strategy Becomes Sub-Optimal at an Earlier Stage

C0 C1

Q1 - As lower information costs drop the cost of intervention from C0 to C1, the point at which intervention is justified shifts from Q0 to Q1

Q0 Resource pressure (cost of not managing the resource)

241

Rose, Rethinking Environmental Control, supra note 117, at 17; see generally Harold Demsetz, Toward a Theory of Property Rights, 57 A M ECON. REV. 347 (1967) (discussing when control systems emerge). 242

Source: refinement on Rose, Rethinking, figure 1, at 6.

79

Where the pressure on a resource has gone past the point where doing nothing is optimal, information systems and the “indicators” they make possible can highlight the changed circumstances and make it easier to confront the entrenched expectations of those who have been taking advantage of belowcost resources.243 Remote sensing and computer tracking, for instance, makes it easier to monitor the size of fish stocks and focus fishermen on the need for conservation regimes to prevent a fisheries collapse. Likewise, satellite imaging makes deforestation much easier to track, making more vivid the harmful effects of those indiscriminately clearing the land. c. Assumption of Rough Reciprocity (“live and let live”) A corollary of the theory that abundant resources need not be managed is the notion that if pollution harms or resources consumed are roughly reciprocal, then the cost of intervention to internalize externalities is not justified. In particular, if A is causing ten units of harm to B with the smoke from his coal furnace, but B is causing ten units of harm to A with the smoke from his coal furnace, it makes little sense (or so the theory goes) for A to be required to pay B for the ten units of harm only for B to pay A the same amount back again. The calculus that intervention is not justified seems especially strong when the harms are small and the cost of taking action to internalize the externality is not. The reciprocity rationale is, however, a crude rule and increasingly inappropriate in our modern day. First, the assumption of roughly equal harm across users is often wrong. In a few instances, such as breathing, the use of the resource (air) and the emissions released (carbon dioxide) are roughly equal across all individuals. But in other cases some people use far more than their share of a collective resource or emit far greater than average pollution. Those who consume a much smaller fraction of common resources or emit less than others are disadvantaged by a presumed equality of impacts. Tailpipe controls under the Clean Air Act limit but do not eliminate vehicle emissions. But there has been no serious attempt to require compensation for the not-insubstantial residual harm from exhaust. The harm

243

ROBERT SUGDEN, THE ECONOMICS OF RIGHTS , COOPERATION, AND W ELFARE (1986); Robert C. Ellickson, Bringing Culture and Human Fraility to Rational Actors: A Critique of Classical Law and Economics, 65 CHI.-KENT L. REV. 23, 23 (1989) (highlighting the difficulties in overcoming entrenched expectations).

80

caused is assumed to be more or less reciprocal to the extent that everyone in the United States drives a car.244 But if Amory drives a highly fuel efficient hybrid car (and sometimes rides his bike) while Betty drives a gas-guzzling Sports Utility Vehicle — their air pollution impacts are by no means roughly equal. Betty’s emissions will certainly be three or four times greater than Amory’s and might be as much as ten times higher. Second, the assumption of rough reciprocity, which focuses our attention on the balance of emissions between Amory and Betty as drivers, distracts us from the real underlying allocative inefficiency. Specifically, when emissions are not fully priced, the consumption by both Amory and Betty of the polluting good (cars) is subsidized relative to other goods (e.g., bicycles). As a result, polluting goods and activities (driving gas-powered cars) are over-consumed relative to non-polluting ones (riding bicycles) or less polluting ones (driving hybrid vehicles).245 If both Amory and Betty were to face the full costs of driving, both of them might choose to drive less, make more efficient use of their cars, or create a demand for vehicles that generate less pollution. Third, and most importantly, as the cost of tracking pollution and pricing externalities falls, it will become economically rational to address environmental harms in a larger number of cases. In fact, as the transaction costs of pollution monitoring approach zero, the number of cases where presumed reciprocity makes sense will shrink towards zero. Ultimately only in cases where harms are truly reciprocal and de minimis will it be efficient to ignore them. d. Small Equals Zero Beyond disregarding presumed reciprocal harms, the existing environmental regime often treats small harms as unimportant and thus not worth bothering to address.246 But we know that there are a

244

KRIER & URSIN, POLLUTION AND POLICY, supra note 52.

245

BAUMOL & OATES , ENVIRONMENTAL POLICY, supra note 59, at 71-79 (describing the limitations of free market pricing systems wherein the costs of environmental externalities are not reflected in the costs of consumer products). 246

See Sunstein, Probability Neglect, supra note 94 for a discussion of the tremendous insensitivity to low probability events, explained by both a human inability to comprehend differences among low probabilities, and a psychic desire to establish “safety thresholds,” thereby eliminating anxiety over real, but low probability risks.

81

large set of pollution sources that are individually so tiny that one might imagine that their modest contribution to the overall environmental quality could safely be ignored. When these small sources are aggregated, however, the impact can be very big. Thus, if Betty’s car were the only one on the roads, it would not merit any attention. But alongside 200 million other vehicles on America’s highways and byways, the cumulative air pollution from vehicles looms very large, no matter that the harm from each is insignificant.247 The cumulative impact of small-source emissions today represents the bulk of the remaining pollution problem in many areas.248 As with other cases where a do nothing response has been presumed to be appropriate, the logic for ignoring small harms collapses as the cost of tracking and internalizing them drops. As transaction costs fall the efficiency gains from internalizing the externality will outweigh the administrative burden in more and more circumstances.

VI. DRIVING ENVIRONMENTAL INFORMATION INTO THE DECISIONMAKING R EALM Some of the information gaps that plague efforts to optimize environmental protection will naturally shrink over time. Our environmental knowledge base is growing steadily. Although the world is dynamic and new challenges are constantly emerging, knowledge is cumulative.249 Ongoing investments in environmental data and analysis broaden the information base for future decisionmaking. Advances in allied fields (statistics, biology, chemistry, epidemiology, meteorology, etc.) constantly strengthen the foundations for environmental decisionmaking. Technical, scientific, and analytic constraints to sound 247

Americans cumulatively drive more than 2 trillion vehicle-miles per year. Federal Highway Administration, Traffic Volume Trends, at http://www.fhwa.gov/ohim/tvtw/tvtweb.pdf. (last visited Oct. 14, 2000). 248

EPA, National Air Quality and Emissions Trends (1997), available at www.epa.gov/car/aqtrnd97/toc; EPA, National Air Quality and Emissions Trends (1998), available at www.epa.gov/oar/aqtrends.html/toc.html; see also EPA, National Air Pollutant Emission Trends: 1900-1998 (Mar. 2000), available at http://www.epa.gov/ttn/chief/trends/trends98/. 249

In some cases, theories unravel over time as we determine that they were based on faulty analysis. But more often, knowledge is refined as more precise data and thinking becomes available. Thus, the technical gap trends downward over time. See, e.g., Robert Wright, Mr. Order Meets Mr. Chaos, FOREIGN POL’Y, May-June 2001, at 50 (history has a “direction”). But see THOMAS HOMER-DIXON, THE INGENUITY GAP (2000) (arguing that the world is getting more complex and thus that the difficulty of establishing a sufficient information foundation for decisionmaking is worsening rather than improving).

82

environmental decisionmaking will therefore tend to soften over time even without any conscious policy aimed at gapfilling. In fact, one of the remarkable stories behind the Information Age is how much environmentally relevant data and knowledge is being generated and shared without any plan, governmental mandate, or structured set of incentives to promote innovation. Nevertheless, the design of the legal system and the incentives (or disincentives) it provides for data generation and analysis can affect the pace at which information gaps are filled. In many instances, the current policy structure discourages those who are positioned to generate critical information from doing so. What interest does the fisherman have in collecting data on by-catch? Why would a factory manager look for and report on environmental risks she isn’t required to track? Motivation thus matters. From a policy perspective, more emphasis should be given to luring or driving data and accompanying analysis into the marketplace, the regulatory process, and the hands of consumers.250 In developing incentives for information production, it is useful to understand why information that would be useful to decisionmakers is not now available. In this regard, the thinking to date has been dominated by an economics perspective that: (1) focuses narrowly on internalizing externalities and (2) sees information in terms of whether it is a public good or a private one and whether it is rival in consumption or not. 251 These dichotomies are of limited analytic value in the environmental domain. A. Beyond Externalities In the classic externality situation, B benefits from externalizing environmental costs onto A or the community-at-large. Benton saves money in his pig farming operation if he can dodge the burden of

250

Ironically, the recent trend has been to disinvest in data production. The Congress abolished its own Office of Technology Assessment in 1995. Legislative Branch Appropriations Act 1996, Pub. L. No. 104-53 §112, 109 Stat. 468, 526 (1995). Congress has also refused to fully fund the National Biological Survey. Charles Oliver, A New Push on the Environment?, INVESTOR ’S BUS . DAILY, Dec. 5, 1996, at A1 (noting the Clinton Administration’s efforts to use administrative regulations to continue attempts to catalogue plant and animal species in the U.S. in light of the congressional refusal to fund the National Biological Survey). The National Biological Survey (later renamed the National Biological Service) was terminated in a 1996 spending bill and transferred in part to the Geological Survey. Balanced Budget Downpayment Act I, Pub. L. No. 104-99, §123, 110 Stat. 26, 33 (1996). 251

Graciela Chichilnisky, The Knowledge Revolution, 7 J. INT ’L TRADE & DEV. 39, 41 (1998) [hereinafter Chichilnisky, Knowledge Revolution]; A DAM BRANDENBURGER & BARRY NALEBUFF, COOPETITION (1996) (explaining the dynamics of information generation and use).

83

addressing the effluent and odors that emanate from his sty.252 There are, however, a set of environmental problems that arise from production or consumption inefficiencies or mistakes. In such cases, the problem can be understood as a direct information failure. B does not need to be forced by law or regulations to internalize his externalities. He has an incentive to fix the problem regardless of the impacts on A or others. Think, for example, of cases where B’s factory equipment is outdated and inefficient—and incidentally, highly polluting. If the plant were modernized, efficiency benefits would flow to the factory’s bottom line. Those downwind of the polluting production process would also benefit as increased resource productivity translates into reduced emissions.253 In some circumstances, B fails to act because he believes he has better ways to deploy limited investment capital. But in many other cases, B’s inaction derives from inertia or a lack of information about the better environmental options available. Such circumstances differ from the usual externality case because self-interest, at least over time, should drive B to correct his “mistake” and adopt the environmentally preferable technology, product, or production process.254 Pressure from those suffering from B’s emissions, or government officials acting on their behalf, need not be brought to bear to motivate action. In any case, environmental policy in this context should center on providing information about better practices and potential efficiency gains rather than narrowly on internalizing externalities.255 A great deal of attention has been paid in recent years to opportunities for such “win-win” environmental gains.256 A vigorous debate rages over how often polluters obtain net benefits from changing 252

Aldred’s Case, 77 Eng. Rep. 816 (1611).

253

Michael E. Porter & Claas van der Linde, Toward a New Conception of the Environment-Competitiveness Relationship, 4 J. ECON. PERS. 9 (1995). 254

Esty & Porter, Industrial Ecology, supra note 37, at 36-40 (discussing resource productivity opportunities available to companies). 255

Progress is likely only “over time” because of the need to depreciate existing capital investments and the “lumpiness” of investments in new plant and equipment. To speed up the transition process, the government should highlight the better options and efficiency gains available. To ensure an optimal degree of investment in pollution control, B’s decision process should be structured to reflect externalized pollution harms. But a pure “cost internalization” strategy will not highlight the potential gains within the enterprise from higher-efficiency, less-polluting equipment and will systematically result in less new capital investment. 256

A MORY LOVINS , HUNTER LOVINS , & ERNST ULRICH VON W EIZSÄCKER, FACTOR FOUR : DOUBLING W EALTH – HALVING RESOURCE USE (1997) (arguing that substantial efficiency improvements are often available); A L GORE, EARTH IN THE BALANCE : ECOLOGY AND THE HUMAN SPIRIT (1992) (describing a world full of win-win opportunities).

84

their own behavior.257 The truth is that investments in pollution control sometimes but by no means always pay off economically as well. Nonetheless, the mistake and inefficiency category of environmental problems deserves particular attention in the Information Age context because easier and cheaper dissemination of information about the existence of better technologies and production practices offers the potential for motivating action. B. Environmental Information as a Public Good

A number of economists have analyzed environmental information as a matter of private versus public goods. Chichilnisky asserts, for instance, that knowledge is a privately produced public good that is non-rival in consumption.258 But she overlooks the fact that a great deal of environmental data and information are publicly generated. Dozens of reports are put out each year by the U.S. Environmental Protection Agency (EPA), the Interior Department, and other government agencies. Environmental knowledge also emanates from public research centers such as EPA laboratories, the National Academy of Science, or the National Institutes of Health. Government- supported research, such as work funded by the National Science Foundation at academic institutions, produces a vast store of additional data and analysis. Non-profit think tanks and private research centers also generate significant data, information, and learning with a degree of public funding. It is therefore a significant over-simplification to see environmental knowledge as privately produced. A good bit of privately generated environmental knowledge is , furthermore, not shared publicly but rather guarded for commercial advantage. Such information cannot be analyzed as a traditional public good. Consider, in this regard, the cutting-edge, pollution-reducing seeds developed by biotechnology companies. These products are not “rival” in consumption the same way that an apple can be eaten by

257

Michael E. Porter, America’s Green Strategy, SCIENTIFIC A MERICAN August 1991 at _____ (advancing the “Porter Hypothesis,” which suggests that environmental investments may result in competitiveness gains); Michael Porter & Claas van der Linde, Green Competitiveness, 73 HARV. BUS . REV. 120 (1995) (discussing how environmental rigor can generate innovation benefits); but see Noah Walley & Bradley Whitehead, It’s Not Easy Being Green, 72 HARVARD BUS . REV. 3 (1994) (suggesting that win-win gains are very limited in the real world). 258

Chichilnisky, Knowledge Revolution, supra note 251, at 42 (suggesting that knowledge is a privately produced public good).

85

only one person. But neither are they non-rivalrous in the strong sense of traditional public goods like national defense, which is not only non-rival but also “non-excludable.” Access to knowledge can be and often is controlled for economic advantage. Environmental data, information, and knowledge need to be understood, therefore, in a more rounded form and not as pure public goods that are available for consumption by all. A more useful analytic taxonomy focuses on: (1) whether the data, information, and analysis needed for a functioning market or sound policymaking exist and (2) whether this knowledge is viewed as a strategic asset. A further distinction can be drawn between information that is held close to keep it out of the hands of the government versus that which is not shared because of its value to competitors. 1. Critical Environmental Information Does Not Exist In some cases, data and information that would improve environmental decisionmaking exist but are not accessible to decisionmakers. But in many other cases, essential information or knowledge is simply not available. These gaps can be traced to (a) technical and analytic limitations which derive from cost barriers; (b) market failures; and (c) institutional shortcomings. a. Technical and Cost Barriers As discussed earlier, the optimal specificity of regulation is a fundamental issue in the environmental realm. Sometimes we know what data we would like to have to make a policy decision, but we choose not to obtain the information because the cost of acquisition would not be justified by the anticipated gains in policy precis ion. It would be technologically feasible, for example, to stop each car on the road and measure its particulate emissions. From a policy control perspective, the data generated would be enormously useful. But the cost of undertaking such a detailed monitoring program (in money, time, and driver inconvenience) has been seen as prohibitively expensive — especially in comparison to the expected benefits. In other cases, the information production process is simply slow. Precise levels of pesticide safety have not yet been established for many food products because exposure and risk analysis take time and money. When cost is the central issue, putting incentives in place for data generation

86

emerges as a critical policy focus.259 In too many cases, those best positioned to generate information have no reason to produce what is needed and perhaps even an incentive not to create or release the data.260 In these circumstances, intervention is required to change the information-generation calculus of those who produce the products in question — who also often happen to be the least-cost information generators. In many respects the environmental information gap is a cost/benefit issue. In some cases, the cumulative (social) costs of generating needed data and analysis exceed the social benefits so the work does not get done. In other cases, the private costs of generating (or sharing) critical information exceeds the private benefits, so those with the data or analysis (often a polluting or resource consuming entity) do not come forward. Where the social costs of obtaining information exceed the benefits, the pace at which new data and relevant knowledge is produced can be affected by the choice of regulatory strategies. Specifically, incentives can be developed to lower the costs of information gathering, increase the benefits, or shift the cost/benefit calculus to a basis that reflects social rather than private costs and benefits. In some circumstances, scale economies can be achieved by centralizing the analytic process. In this regard, some observers have called for a government-led environmental “super study” to fill critical data and information gaps.261 Others argue that competition and decentralization of the data gathering, information generation, and analytic processes will generate more useful foundations for decisionmaking. 262 The competition between rival groups to map the human genome demonstrates this potential. 263 259

Lyndon, Information Economics, supra note 30.

260

Wagner, Choosing Ignorance, supra note 31; IAN A YRES & JOHN BRAITHWAITE, RESPONSIVE REGULATION: TRANSFORMING THE DEREGULATION DEBATE 45 (1992). 261

Lyndon, Information Economics, supra note 30, at 1839-42.

262

Esty, Revitalizing, supra note 21, at 605-45 (assessing the benefits of regulatory centralization versus decentralization more generally) 263

Frederic Golden et al., The Race is Over, TIME, July 3, 2000, at 18.

87

Whether cooperation or competition and a lead role for government or the private sector will work best must be determined on a case-by-case basis depending on the structure of the issues and sectors in question. 264 Ensuring that the opportunities for improved information generation are systematically explored in the regulatory context will require devoting more effort to identifying the least cost generator of data and analysis and to designing environmental rules to deliver an optimal degree of relevant information to decisionmakers at the lowest possible cost. A number of proposals might be considered in this regard. One idea would be to require an “information gap analysis” as part of the regulatory process. This could be done by Executive Order, as a supplement to the existing requirements for regulatory impact and cost-benefit analyses. Under such a process, the EPA would be required when promulgating new Clean Air Act regulations to factor into its rulemaking process some overt consideration of (1) what information is needed to make sound air pollution policy decisions, (2) what data and analysis is missing or incomplete, (3) who is best positioned to generate this information, and (4) what incentives could be adopted to ensure that the requisite data and information gets collected and analyzed. The Office of Information and Regulatory Affairs (OIRA) in the White House Office of Management and Budget (OMB) might be asked to take the lead in reviewing the information analyses produced by the regulatory agencies and in identifying opportunities to drive information into the decisionmaking process. OIRA might also be charged with identifying over-arching information gaps and possible strategies for filling them. A further refinement might entail amendment of the Administrative Procedure Act to require that consideration be explicitly given to information gaps and how to fill them. Making the data and analytic foundation for decisionmaking part of the Notice and Comment process would engage a broader set of actors in thinking through the knowledge generation questions that are part of any regulatory action. 265 264

DANIEL C. ESTY & DAMIEN GERADIN, REGULATORY COMPETITION AND ECONOMIC INTEGRATION: COMPARATIVE PERSPECTIVESE (2001), at xxi-xxv (discussing the cooperation versus competition dynamic). 265

In a similar vein, Graham and Wiener have called for a more holistic public health and environmental regulatory process based on better data and analysis and centered on a structured evaluation of risk tradeoffs. See W IENER & GRAHAM & W IENER, RISK VS . RISK, supra, note 19.

88

Such a requirement might help to smoke out critical information. At a minimum it would help to clarify who the “least cost information generators” might be. b. Market Failures In some cases, the gains from generating environmental information exceed the costs from a society-wide perspective, but the distribution of these benefits and costs means that useful data and analysis does not get produced or shared. For instance, while society would benefit from a better understanding of the range of endangered species, the landowners on whose property these species might reside have no incentive to develop the requisite data. 266 From the private landowner’s perspective, reduced uncertainty creates risk and offers little advantage. Notably, the landowners recognize that if the species is determined to be on their property, they face restrictions on the use of the land, new regulatory burdens, and lower property values.267 Thus, they have little interest in surveying the species present on their land or cooperating with governmental mapping exercises. Such asymmetries between the costbearers and the beneficiaries of environmental information generation lead to chronic under-investment in relevant data and analysis. 268 To enhance decisio nmaking in the context of endangered species tracking, wetlands mapping, or any number of other circumstances where data is under-produced, this calculus must be changed. One possibility would be payments to property owners whose land contains resources that generate

266

Michael Bean has spelled out how Habitat Conservation Plans under the Endangered Species Act might be restructured to get landowners to disclose the presence of endangered species on their property. Michael J. Bean, The Endangered Species Act and Private Land: Four Lessons from the Past Quarter Century, 28 ENVT ’L. L. REP . 10701 (1998). 267

Jacqueline Brown, Preserving Species: The Endangered Species Act Versus Ecosystem Management Regime, Ecological and Political Considerations, and Recommendations For Reform, 12 J. ENVTL. L. & LITIG. 151 (1997) (noting that perverse incentives arise under the existing regulatory s cheme); Maura Dolan, Nature at Risk in a Quiet War, L.A. TIMES , Dec. 21 1992 at A1 (describing various instances of private citizens killing endangered species or deliberately destroying habitat). 268

See Lyndon, supra note 30, at 1810-817 (discussing disincentives to production of chemical toxicity and exposure data).

89

community benefits. 269 Subsidizing positive externalities would reverse the current incentive to hide such resources and to fight information production efforts.270 In these cases and others where the information needed for environmental decisionmaking has the characteristics of a public good, the underlying market failure could be corrected by governmental investments in information production. Data generation and analysis might therefore be seen as a priority role for the Environmental Protection Agency, the U.S. Fish and Wildlife Service, and other pollution control and resource management agencies.271 Where communal resources are going under-protected at the local, state, regional, and global scales, similar investments in environmental information would also be appropriate. But the government may not be the best information provider. It may make sense for the government to pay others to generate the information needed for sound environmental decisionmaking. Alternatively, government mandates to the least-cost information provider may be appropriate, especially if this party is also the one creating the harm in issue. In the pollution context, where the externalities are largely negative, making the polluting enterprises generate and release data on their harms will often make sense. In natural resource management, where the externalities will often be positive, it may be that the government should compensate those who are asked to generate needed data, inducing landowner cooperation.

269

Andrew G. Frank, Note, Reforming the Endangered Species Act: Voluntary Conservation Agreements, Government Compensation and Incentives for Private Action, 22 COLUM. J. ENVTL. L. 137, 145-46 (1997) (noting that current ESA regulations give landowners an incentive to destroy or hide species on their land or develop land before ESA regulations are in place). 270

Carol Rose suggests a variety of incentives to induce environmental cooperation from private property owners. Many of these tools, such as tax breaks, might be used to promote environmental information generation. Carol M. Rose, Property Rights and Responsibilities, in THINKING ECOLOGICALLY, supra note 44. 271

Ironically, the trend has been in the opposite direction. During the 1990s significant disinvestments in data generation occurred including the elimination of the Office of Technology Assessment and cutbacks in funding for the National Biological Survey. See Robert L. Glicksman & Stephen B. Chapman, Regulatory Reform and (Breach Of) The Contract With America: Improving Environmental Policy or Destroying Environmental Protection?, 5 KAN. J. L. & PUB . POL’Y 9 (1996) (criticizing the elimination of the Office of Technology Assessment and efforts to eliminate the National Biologic Survey); Dennis Hayes, Environmental Law and Millennial Politics, 25 ENVTL. L. 953, 958 (1995) (criticizing Congressional efforts to cut knowledge generation).

90

At the same time, environmental disinformation needs to be disciplined. 272 Governments need to be vigilant about policing environmental claims.273 If misinformation and false promises of biodegradibility, recycled content, recyclability, etc. are not prosecuted, consumer confidence in environmental claims will falter and the strength of market signals will be muted. 274 Where systems now exist to monitor the truth of claims, such as through the U.S. Federal Trade Commission, the review process may need to be strengthened as information-based regulation becomes more central to society’s regime of environmental protection. In particular, procedures will be required to screen against information overload or disinformation designed to confuse the consumer or overwhelm the policy process.275 Ultimately, a set of companies may emerge who act as information filters and content evaluators, providing consumers (presumably over the Internet) with reliable information on the environmental attributes of products.276 But while electronic commerce is still in its infancy, some interim governmental support for the integrity of this emerging marketplace may be needed. c. Legal and Institutional Failures Some statutes and regulations actively reward ignorance. U.S. pesticide laws and regulations , notably, the Federal Fungicide, Insecticide, and Rodenticide Act or FIFRA277 , allow corporations to keep 272

David Roe, STARTING BLOCKS nn.21 & 24 (unpublished Environmental Defense Strategy Paper, 2000), available at www.edf.org/wip/startingblocks (highlighting disinformation effects). 273

Without government regulation, misleading or false claims of environmental product attributes are likely to proliferate, rendering environmental labeling meaningless. In 1991, for example, the State of Washington reached a settlement with Mobil Chemical Company regarding advertising Hefty trash bags “degradable” claims. The Hefty packaging seemed to promote the product as environmentally-sensitive and biodegradable although the bags were actually unlikely to decompose in land fills where there is no sun exposure. June Camille Bush Raines, The Green Giant: Environmental Marketing Claims, 45 OKLA. L. REV. 689, 699 (1992). 274

See REINHARDT , DOWN TO EARTH, supra note 232, at 39-43 (2000).

275

Over time, those who say inaccurate or foolish things tend to lose their audience. But it may be useful to improve the workings of the intellectual marketplace by facilitating efforts to squeeze out purveyors of misinformation. 276

An example of such a phenomenon is the development of technologies designed to filter out violent or sexually explicity television shows or websites. See Jack M. Balkin, Media Filters, The V-Chip, and the Foundations of Broadcast Regulation, 45 DUKE L.J. 1131 (1996); C. Dianne Martin and Joseph M. Reagle, The Jurisprudence of Rating Symposium, An Alternative to Government Regulation and Censorship: Content Advisory Systems For the Internet, 15 CARDOZO A RTS & ENT . L. J. 409 (1997). 277

7 U.S.C. §136a (2000) (requiring pesticide manufacturers to present “a full description of the tests made and the results thereof” to register a pesticide).

91

their products on the market until regulators prove them to be harmful. Companies have no affirmative duty to demonstrate the safety of their products and thus little incentive to do detailed risk analyses until the government raises a question about a particular product.278 Beyond FIFRA, the Toxic Substances Control Act (TSCA),279 Resource Conservation and Recovery Act (RCRA),280 the Comprehensive Environmental Response, Liability, and Cleanup Act (CERCLA) or “Superfund,” 281 and a number of other laws are structured so that potential harm causers have limited interest in discovering or carefully examining the risks they may be creating. Such disincentives for information generation represent a critical regulatory failure. Companies have no incentive, furthermore, to look for harms in areas where there are no laws or regulations. Many areas of potential risk, such as indoor air pollution, go unattended because there is no reason to invest in harm identification, impact evaluation, mitigation or abatement. The problem again can be traced to the structure of environmental laws – particularly the limited coverage of the current patchwork of statutes. Strategies exist, however, for reversing the incentives for ignorance. First, a number of laws, such as FIFRA, would benefit from a strategy of “burden shifting.” California’s “Proposition 65” shows the potential in this regard. 282 Under this voter-approved California toxics law, the burden of proving safety shifts from the government to the producers of products that may be carcinogenic or expose the public to reproductive hazards.283 Rather than allowing manufacturers to sell their wares until the 278

PERCIVAL ET . AL., ENVIRONMENTAL REGULATION , supra note 23, at 434-435.

279

15 U.S.C.A. §§ 2601 – 2692 (West Group, 2002).

280

42 U.S.C.A. §§ 6901 – 6992(k) (West Group, 2002).

281

42 U.S.C.A. §§ 9601 – 9675 (West Group, 2002).

282

The Safe Drinking Water and Toxic Enforcement Act of 1986, Cal. Health & Safety Code (CH & SC) ch. 6.6, Secs. 25249.5 et seq, (West 1986); see also www.prop65kit.org (site last visited May 6, 2003). 283

Clifford Rechtschaffen, The Warning Game: Evaluating Warnings Under California’s Proposition 65, 23 ECOL. L. Q. 303 (1996) (explaining how Proposition 65 works); David Roe, An Incentive-Conscious Approach to Toxic Chemical Controls, 3 ECON. DEV. Q. 179, 179-182 (1989) [hereinafter Roe, Incentive-Conscious]. See also Michael Freund, Proposition 65 Enforcement: Reducing Lead Emissions in California, 10 TUL. ENVTL. L.J. 333 (1997)

92

government proves they are unsafe, Proposition 65 puts the burden on the company to warn the public about exposure risks or to prove that their products are safe.284 Backed by the threat of significant legal penalties for the marketing without a warning of any product creating a greater than 10-5 additional risk of carcinogenity, the California law transforms the incentives for data generation. 285 By placing the safety burden on industry, the California law has generated a vast quantity of chemical exposure data.286 From “White Out” correction fluid to the lead solder in tuna fish cans, dozens of products have been taken off the market or reformulated to reduce their public health risks.287 A related approach involves the careful structuring of “default rules” and legal presumptions. A presumption against the party in the best position to produce relevant data creates an incentive to resolve uncertainties. For example, the existing rules governing pesticides permit the marketing of new products unless the government is able to demonstrate a safety problem. 288 Creating a “duty to test” or a presumption that any new product must have its safety demonstrated before being allowed into the marketplace would create incentives for producers to generate data and thereby reduce uncertainty. Applied more broadly, such presumption and default rules might generate significant new investments in environmental data as well as testing protocols and equipment. All of this information would permit a much more refined and careful process of environmental standard setting. Fundamentally, if uncertainty

284

This approach can be seen as an application of the Precautionary Principle. Daniel Bodansky, Scientific Uncertainty and the Precautionary Principle, 33 ENV’T . 4, 5 (1991); GRAHAM & W IENER, RISK VERSUS RISK, supra note 19. 285

The Safe Drinking Water and Toxic Enforcement Act of 1986, Cal. Health & Safety Code §§25249.6-25249.10 (Deering 2001). 286

David Roe, Toxic Chemical Control Policy: Three Unabsorbed Facts, 32 E. L. R. 10232 (2002) (spelling out how Prop 65 works); David Roe, Ready or Not: The Coming Wave of Toxic Chemicals, 29 ECOL. L. Q. 623 (2003) (highlighting weaknesses in the federal approach to toxics regulation). 287

PERCIVAL ET AL., ENVIRONMENTAL REGULATION, supra note 23, at 525 (noting the removal of a carcinogenic chemical from Kiwi water-proofing spray, and lead solder from cans containing food products). See also Roe, Incentive-Conscious, supra note 283. Roe notes that under TSCA, the federal government had “applied quantified testing limits” to only 17 chemicals between 1976 and 1988. Id. at 185 n.27a (citing 40 C.F.R. §799.1) (California managed to draw bright lines for more chemicals in the first twelve months of the Proposition 65 era than the federal government had managed to accomplish, under the supposedly omnibus Toxic Substances Control Act, in the previous twelve years). 288

7 U.S.C. §136a (2000).

93

were to carry a clear price tag, data on environmental risks would be generated as companies seek to reduce their cost burdens. Accounting rules may also discourage information production. Under current Federal Accounting Standards Board rules, for example, environmental liabilities, such as potential responsibility for Superfund cleanups that are “contingent” and “unquantified” can be reported on in passing in audit statements.289 In contrast, more well-documented liabilities require fuller explanations and the setting aside of reserves. Similarly, SEC disclosure rules sometimes discourage companies from knowing the full extent of their environmental liabilities because more complete information triggers broader disclosure requirements.290 Revised rules could change these incentives.291 Raw information on environmental harms and mitigation options is sometimes available but not systematically collected, analyzed, or converted into a form that supports sound decisionmaking. Gaps of this sort often persist as a result of neglect more than intent. For example, a good bit of corporate information on environmental conditions, risks, control technologies, and other variables is not publicly available because companies see little reason to collect it or share it. Some companies, as part of their outreach to local communities and NGOs, have begun to issue environmental or sustainability reports that present aggregate data on their pollution control and resource management performance.292 But much of the most useful information—especially on factory or facility level results—remains buried. A shift

289

Financial Accounting Standard No. 5 requires tabulation of a contingent liability if it is probable to occur and the amount of loss can be reasonably estimated. Although there have been recent efforts to clarify how this standard applies to environmental liabilities, (including specific guidance outlined in Staff Accounting Bulletin No. 92), companies are allowed to make their own determination of whether an environmental liability must be reported under this standard with very little oversight. Susan Millington Campbell & Carol E. Remy, Disclosure Requirements and Securities Laws, N.Y. L. J., Oct. 30, 1997, at S4. 290

Michael Gollub, Reducing Uncertainty in Environmental Disclosure: Why the Securities and Exchange Commission Should Return to the Basics, 4 ENVTL. L. 311 (1998) (providing an overview of SEC rules on disclosure of environmental liabilities). 291

Steven L. Bray, Sealing the Cracks in the SEC’s Disclosure Rules: A Risk Communication Approach, 18 U. PENN . J. INT ’L. ECON. L. 655 (1997) (highlighting some needed reforms). 292

Gary S. Guzy, Reconciling Environmentalist and Industry Differences: The New Corporate Citizenship ‘Race To the Top’? ” 17 J. LAND USE & ENVTL. L. 409, 414 (2002) (noting that fifty percent of the world’s largest companies, the Fortune 100, now prepare environmental sustainability reports).

94

toward a more empirical and quantitative approach to environmental protection could create a logic for tabulating and releasing this information. 293 A regulatory shift toward performance targets (away from technology mandates) might spur production of more refined environmental indicators and reporting. Corporate information incentives could be restructured by a shift from a command and control regulatory strategy to a “command and covenant” model. As Elliott suggests, such an approach puts the burden on corporations to measure and report on the public health or ecological harms they are causing and to then negotiate reduction targets.294 Additional administrative failures may arise because the entity best positioned to undertake environmental decisionmaking is not the same entity that is best positioned to do environmental data gathering or processing. As noted earlier, a number of environmental choices currently made by federal or state regulators might better be devolved to local communities or individual consumers. Where externalities do not extend beyond a relatively small jurisdiction, local or state decisionmakers may be better positioned than distant federal regulators to understand all of the relevant circumstances and make appropriate tradeoffs. But, given the scale economies in some dimensions of environmental information generation, a prerequisite to successful devolution of environmental decisionmaking might well be the creation of a National Institute for the Environment which would do broad-based data collection and analysis.295 This reality argues for more emphasis on hybrid institutional structures and multi-tier regulatory approaches which divide information gathering and analysis responsibilities among levels of government. 293

Daniel C. Esty, Toward Data-Driven Environmentalism: The Environmental Sustainability Index, 31 ENVIRO. L. REP . 10603 (2001) (discussing the possibility of a mo re empirical and quantitative approach to environmental protection). 294

E. Donald Elliott, Toward Ecological Law and Policy, in THINKING ECOLOGICALLY, supra note 44. See also ENVIRONMENTAL CONTRACTS : COMPARATIVE APPROACHES TO REGULATORY INNOVATION IN THE UNITED STATES AND EUROPE (Eric Orts & Kurt Deketelaere eds., 2001) [hereinafter ORTS & DEKETELAERE, ENVIRONMENTAL CONTRACTS]; Elliott & Charney, Bigger Bubbles, supra note 91 (explaining how this approach creates incentives for information generation). 295

A number of attempts have been made to create a National Institute for the Environment (NIE). For example, the 1997 “Sound Science for the Environment” bill called for the creation of the NIE as an independent organ of the National Science Foundation. H.R. 2914, 105th Cong. (1997); see also, David E. Blockstein, New Environmental Institute Needed to Focus on Emerging Issues, 12 FORUM A PP . RES . & PUB. POL. 133 (1997).

95

2.

Information Withheld In some cases, environmental information exists, but those who have the data do not share it.

Companies may not be forthcoming with environmental information because they see themselves in an adversarial relationship with regulators. In other cases, businesses may think that their superior environmental data, analysis, or technologies are a strategic asset and a potential source of competitive advantage. a. Antagonism Toward Government As Ayres and Braithwaite have argued, if our regulatory approach were less contentious, companies would be much more likely to generate and bring forward information about their polluting activities and opportunities for the control of emissions. 296 The call for a shift in regulatory focus to command and covenant emanates from the same spirit. 297 More broadly, the litigious nature of U.S. society has led to a strong bias in the business and legal worlds against generating data and analysis that might later have to be produced in the course of litigation or shared with regulators. As noted above, one way around the strategic withholding of information is through environmental “contracting.”298 If entities within the regulated community were allowed to propose programs to reduce environmental risks, they would have an incentive to generate and share data. Environmental contracting as a strategy to overcome information failures and to promote better environmental protection at a lower cost has long been used by the Dutch.299 Similar approaches have been tried in the United States on a very modest scale. In particular, EPA’s XL initiative promised a

296

IAN A YRES & JOHN BRAITHWAITE, RESPONSIVE REGULATION: TRANSFORMING THE DEREGULATION DEBATE 45 (1992).

297

E. Donald Elliott, Toward Ecological Law and Policy, in THINKING ECOLOGICALLY supra note 44.

298

DEKETELAERE & ORTS , ENVIRONMENTAL CONTRACTS , supra note 294.

299

Richard B. Stewart, Environmental Regulation and International Competitiveness, 102 YALE L. J. 2039, 2090-91 (noting that in the Netherlands, “[e]ach industry is allocated a designated share of the required reductions and improvements. The responsible government authorities and a number of industry groups . . . have signed or are currently negotiating contracts in which the industry agrees to achieve the overall targets assigned to it. In return, the government agrees to substitute the contractual arrangements for the pollutant-by-pollutant regulations otherwise applicable”).

96

degree of regulatory flexibility for companies that committed to “beyond compliance” performance.300 The limited success of this effort highlights the need for careful program design and clear statutory authority. 301 A number of other long-established policy practices exacerbate the tension between regulators and the regulated community and stifle information availability. Notably, the threat of prosecution for shortcomings identified in company environmental audits has chilled interest in many corporations in measuring and documenting the environmental problems they face.302 A “safe harbor,” which gives companies a clear incentive to undertake rigorous environmental self-evaluations would therefore be useful. 303 b. Competitive Positioning Environmental information may be held back not because a business sees itself in an adversarial position vis-à-vis the government, but rather because it sees the information as a strategic asset. In such cases, where environmental information is, in effect, intellectual property, optimal information production requires a different tack. Market forces can promote a degree of environmental information sharing, particularly by those with marketable pollution control procedures, equipment, data, or analysis. 300

Dennis Hirsch, Bill and Al’s XLent Adventure: An Analysis of EPA’s Legal Authority to Implement the Clinton Administration’s Project XL, U. ILL. L. REV. (1998); see also Jody Freeman, Collaborative Governance in the Administrative State, 45 UCLA L. REV. 1 (1997) (looking at collaborative approaches to environmental protection). 301

Lawrence E. Susskind and Joshua Secunda, The Risks and the Advantages of Agency Discretion: Evidence From EPA’s Project XL, 17 UCLA J. ENVTL. L. & POL’Y 67 (examining the legal and institutional obstacles facing EPA’s XL program); Thomas E. Caballero, Project XL: Making It Legal, Making It Work, 17 STAN. ENVTL. L.J. 399 (1998) (questioning EPA’s legal authority for project XL); Bradford C. Mank, The Environmental Protection Agency’s Project XL and Other Regulatory Reform Initiatives: The Need For Legislative Authorization, 25 ECOLOGY L.Q. 1 (1998) (same). 302

FOREST REINHARDT , DOWN TO EARTH, supra note 232 (spelling out the risk that corporate environmental audits could aid prosecutors); Frank Friedman, Is This Job Worth It?, 8 ENVTL. F. 23 (May-June 1991) (raising the spector that audits which identify problems that are not fixed could create legal exposure); Susan J. Spicer, Turning Environmental Litigation on its E.A.R.: The Effects of Recent State Initiatives Encouraging Environmental Audits, 8 VILL. ENVTL. L.J. 1, 4 (1997). 303

The EPA Final Policy explicitly prohibits its regulators from employing direct incentives, including reduced enforcement responses, to encourage companies to pursue internal environmental audits. See Incentives for Self-Policing: Discovery, Disclosure, Correction and Prevention of Violation, 60 Fed. Reg. 66, 706 (1995), at III B (1). However, many states have adopted laws or policies that encourage environmental audits by limiting penalties associated with infractions discovered during such audits or offering other forms of regulatory relief. For an overview of these EPA and state policies, see Paulette L. Stenzel, Can the ISO 14000 Series Environmental Management Standards Provide a Viable Alternative to Government Regulation?, 37 A M. BUS . L.J. 237, 267 (2000).

97

Companies with superior emissions control strategies, for example, have an incentive to “share” their knowledge if they can sell their solutions to polluting enterprises in the marketplace. Likewise, if customers value environmental performance, companies will develop environmentally superior products and share information on the benefits. In this context, the most valuable role for government is to help make the market function efficiently — and to highlight the value of good environmental information.

VII. DOWNSIDES OF THE INFORMATION AGE As previous parts of this Article have demonstrated, there are many reasons to be optimistic about the impact of the Information Age on efforts to protect the environment. Many digital technologies are, however, dual-edged swords. This Part explores how and why Information Age breakthroughs might actually harm efforts at pollution control and natural resource management. Some of the risk derives from new pollution threats that may emerge. Other steps backward might arise from unexpected and negative consequences arising within policymaking or decisionmaking processes. A. New Environmental Harms While many observers see Information Age technologies generating new efficiencies and thus reducing environmental impacts, some analysts fear that the Dig ital Age will generate new pollution problems. The introduction of technologies always creates some risk of unleashing new ecological or public health risks.304 For example, the technologies of the Information Age make genetically modified organisms possib le. The potential to produce bio-engineered crops that require less water, fertilizers, and

304

For example, chlorofluorocarbons (CFCs) were the miracle chemicals of the 1950s — used as refrigerants, solvents, propellants, etc. — but were later identified as the culprit in the depletion of the ozone layer. A LAN S. M ILLER & IRVING M. M INTZER, THE SKY IS THE LIMIT : STRATEGIES FOR PROTECTING THE OZONE LAYER (1986).

98

pesticides is considerable.305 But, there also exists a risk of unintended cross-fertilization that creates superweeds or other ecological spillovers.306 Some observers fear that the Information Age will generate new wastestreams. The disposal of old computers and other electronic devices represents a much larger issue today than two decades ago.307 And some of the digital technologies have very short generational life cycles; computers and cell phones are often replaced every two or three years. But the problem of Information Age waste disposal has largely been exaggerated. The part of the wastestream represented by computers and other new technologies is very small.308 In many cases, moreover, these devices replace old information processing technologies (e.g., punch cards) that were more material-intensive and thus a bigger waste issue. A second concern arises from the increased energy demand that might be seen across society as microprocessors are embedded in more and more household and commercial appliances.309 Whether the tendency toward more computer controlled devices will raise or lower energy demand depends on which of two effects predominates. On the one hand, the use of more technologies that require electricity could create a scale effect that increases energy requirements. On the other hand, to the extent that Information Age technologies reduce energy consumption within the devices into which they are placed, there could be a counterbalancing efficiency effect.

305

Kurt Buechle, The Great, Global Promise of Genetically Modified Organism: Overcoming Fear, Misconceptions, and the Cartagena Protocol on Biosafety, 9 IND. J. GLOBAL LEGAL STUD. 283 (2001) (proclaiming the environmental virtues of GMOs); David G. Victor & C. Ford Runge, Farming the Genetic Frontier, FOREIGN A FFAIRS (May/June 2002) (highlighting the promise of GMO-based agriculture). 306

Lakshman D. Guruswamy, Sustainable Agriculture: Do GMOs Imperil Biosafety?,9 IND. J. GLOBAL LEGAL STUD. 461 (2002) (decrying the potential health and ecological hazards of GMOs). 307

See generally, Silicon Valley Toxics Coalition, Exporting Harm: The High-Tech Trashing of Asia (available at www.svtc.org/cleancc/pubs/technotrash.pdf). 308

A recent report by the EPA found that computers, cell phones, and other consumer electronics (including many noninformation related durables, such as televisions and VCRs) account for less than one percent of the total US municipal solid waste stream. Municipal Solid Waste in the United States: 2000 Facts and Figures, EPA Office of Solid Waste and Emergency Response, available at http://www.epa.gov/garbage/report-00/report-00.pdf 309

JOSEPH ROMM, THE INTERNET ECONOMY AND GLOBAL W ARMING (1999) (available at www.coolcompanies.org/paper/.cfn ) (analyzing the energy use of the Information Age economy).

99

To date, there is little empirical evidence to indicate which of these two effects will predominate. But, given the huge scale of potential efficiency gains, and the relatively small amount of electricity that microprocessors draw, it appears likely that, at least over time, the efficiency effect will exceed the scale effect. Other environmental analysts worry that the economic growth stimulated by Information Age technologies could translate into a broad-based increase in consumer buying power and consumption. The prospect of many more cars on the road, energy consuming appliances in houses, and other pollution impacts from increased consumption leads some environmental advocates to fear a “rebound” effect.310 Again, whether this issue represents a real concern turns on whether the scale effects of economic growth turn out to be larger or smaller than the efficiency effects. B. Negative Impacts on Environmental Decisionmaking Information Age technologies may not only create new pollution harms and strains on natural resource management but may also translate into new difficulties in environmental problem solving. 1. Will Information Become Usable Knowledge? Digital breakthroughs will make information more plentiful and less expensive. But this does not guarantee that raw data will be translated into useable knowledge. If the Information Age promise of better environmental decisionmaking is to be realized, a number of pitfalls must be avoided. First, relevant information must get to decisionmakers in a timely fashion. Given the likelihood that many critical environmental choices will be made on a decentralized basis while a good bit of data collection and analysis will occur on a more centralized level, systematic efforts will be required to make sure that environmental data and analysis flows to those who need it.311

310

See, e.g., Frank Gottron, Energy Efficiency and the Rebound Effect: Does Increasing Energy Efficiency Decrease Demand?, Congressional Research Service (July 2001), available at http://www.csa.com/hottopics/ern/01dec/2elect.html (arguing that the “rebound effect,” in which increased efficiency in production translates into lower costs to consumers and thus increased demand, can cause the anticipated reduction in electricity consumption from improved energy conservation to fall short by as much as 40%). See also Todd Litman, Comparing Climate Change Emission Reduction Strategies for Transportation, EPA Working Paper, available at http://yosemite.epa.gov (examining impact of fuel efficiency regulations and other conservation measures on vehicle miles traveled). 311

Douglas Aldrich has identified three requirements for the effective use of information: “the ability to access it, the

100

Second, as more and more data and information is generated, our environmental decision processes face the risk of information overload. As Enriquez has noted, half of all the data ever collected in history has been obtained in the last four years.312 The ability to sift information may soon become as important as the capacity to gather it.313 Environmental decisionmaking processes also remain vulnerable to concerted disinformation campaigns.314 The World Wide Web offers a cascade of crackpot theories, half-baked analysis, and entirely unscientific data sets. Mechanisms to ensure accountability in cyberspace have not yet been fully developed. 315 The free-wheeling spirit of the Internet community 316 creates an added risk of information overload, spamming, and the dissemination of misinformation. 317 Cyberspace represents an intellectual marketplace that currently has few controls on those who would advance information that is wrong or misleading. On-line services may emerge that rate information sources or provide some degree of quality assurance. But the value of the Internet as an

ability to assimilate and analyze it, and the ability to act on it.” DOUGLAS F. A LDRICH, M ASTERING THE DIGITAL M ARKETPLACE 31 (1999). 312

JUAN C. ENRIQUEZ, A S THE FUTURE CATCHES YOU: HOW GENOMICS AND OTHER FORCES ARE CHANGING YOUR W ORK, HEALTH, AND W EALTH (2001). 313

Significant efforts are being made by a number of software companies to produce programs that will sift and sort data and information. Microsoft’s new “Hailstorm” represents one example of this genre of new software. 314

David Roe, Starting Blocks (2000), at notes 21, 24 (available at www.edf.org/wip/startingblocks) (highlighting disinformation risks). 315

Beth Simone Noveck, Designing Deliberative Democracy in Cyberspace: The Role of the Cyber-Lawyer, 9 B.U. J. SCI. & TECH. L. 1 (2003) (detailing informational problems on the Internet and the challenges of enacting accountability and transparency regulations); Edward Lee, Rules and Standards for Cyberspace, 77 NOTRE DAME L. REV. 1275, 1317-21 (2002) (urging Congress to take the lead in promulgating rules to ensure accountability in cyberspace), Anne W. Branscomb, Anonymity, Accountability, and Autonomy: Challenges to the First Amendment in Cyberspace, 104 YALE L.J. 1639, 1646 (1995) (discussing the difficulties of regulating misuses and abuses of informational resources on the Internet from a First Amendment perspective). 316

LAWRENCE LESSIG, THE FUTURE OF IDEAS : THE FATE OF THE COMMONS IN A CONNECTED W ORLD (2002).

317

See, e.g ., Web of Deception: Misinformation on the Internet (Anne P. Mintz, ed. 2002) (alerting readers to the flood of deceptive, misleading, and erroneous information pervading the Web); U.S. Seeks to End What it Sees as Deceptive Email Operation, N.Y. TIMES , Apr. 18, 2003, at C11 (chronicling recent government attempts t o crack down on deceptive Internet practices) and Saul Hansell, Internet is Losing Ground in Battle Against Spam, N.Y. TIMES , Apr. 22, 2003, at C1 (describing the futility of technological and legal efforts to control unwanted solicitation over the Internet, or “spam,” which currently makes up nearly half of all email traffic).

101

information dissemination tool depends, in part, on how quickly mechanisms to discipline the flow of misinformation take hold. 2. Information as a Strategic Asset The Internet promotes quick dissemination of data and information and disregard for intellectual property rights.318 In such a world, best practices and technologies will be able to move quickly to the full range of places where they can be productively employed in response to environmental problems. But lack of respect for intellectual property could, over time, result in reduced incentives for innovation and knowledge generation. Perhaps, in some industries, a judgment will be made that environmental variables should not be treated as a point of competition. In this spirit, companies might share their best practices in response to environmental challenges and agree to compete on other product dimensions. In other cases environmental variables may be so central to a product’s attractiveness that there will be ongoing competition on these factors. 319 3. Network Effects As with any system, there exists a risk that “network effects” will lock in sub-optimal standards and approaches. In the environmental area, the problem of existing technologies become de facto standards that deter innovation has long been understood. 320 Governments can combat this tendency by ensuring that all environmental regulations take the form of performance standards (not technology mandates) and by quickly testing and certifying new methods of meeting requirements. It appears, moreover, that the speed of change brought about by the Information Age may reduce the risk of network effects and encourage continuous thinking about new directions for progress.

318

Mitchell, E-Topia, supra note 8, at 63 (noting the digital information technologies have caused a collapse of the traditional intellectual property approach to knowledge). 319

See, e.g., Niall FitzGerald, Tomorrow’s Wash: Challenges and Opportunities for the Detergents Industry in the 21st Century, Address delivered by the Chairman of Unilever PLC, (Oct. 5, 1998), available at www.unilever.com (casting doubt on whether environmental advances should “remain proprietary, thus limiting the potential benefits to society”). 320

Byron Swift, Barriers to Environmental Technology Innovation and Use, 28 ELR 10202 (1998) (describing the significant and somewhat peculiar barriers to the development and adoption of new environmental technologies, many of which stem from the regulatory system, itself).

102

4. Transparency Traumas Decisionmaking generally improves when more information is available . But the presumption that more complete data and analysis yields better results will not always hold true. In some circumstances, a “Rawlsian veil of ignorance” may actually facilitate agreement on an appropriate policy. 321 For example, climate change negotiators might well find it easier to agree on a strategy for reducing greenhouse gases if they do so in general terms before knowing the precise costs and benefits of emissions reductions in their own countries.322 More generally, transparent policymaking processes may be susceptible to manipulation by special interests unless administrative law and procedures evolve in parallel. The promise of cyberdemocracy with a more fully informed and engaged populace could give way to “spam,” misinformation, and dialogue among the uninformed that diminishes thoughtful deliberation. 323 More opinions being heard could lead to a cacophony and breakdown rather than higher quality decisions.324 And with a multitude of entertainment options only a click away, the Digital Age citizenry may be less rather than more interested in becoming environmentally informed and engaging in participatory decisionmaking. 325

321

See GEOFFREY BRENNAN & JAMES M. BUCHANAN, THE REASON OF RULES : CONSTITUTIONAL POLITICAL ECONOMY (1985) (arguing that fundamental principles are best developed under a veil of uncertainty); ORAN YOUNG, INTERNATIONAL GOVERNANCE: PROTECTING THE ENVIRONMENT IN A STATELESS SOCIETY (1994) (demonstrating how uncertainty can facilitate agreements). 322

Kenneth Arrow, Intertemporal Equity, Discounting, a nd Economic Efficiency, in CLIMATE CHANGE 1995: ECONOMIC CHANGE (Bruce et al., eds., 1996).

AND SOCIAL DIMENSIONS OF CLIMATE 323

Dennis Thompson, James Madison on Cyberdemocracy, in DEMOCRACY.COM: GOVERNANCE IN A NETWORKED W ORLD 36-40 (Elaine C. Kamarck & Joseph S. Nye, Jr. eds., 1999) (noting that “directness” may be advanced at the expense of deliberation) [hereinafter DEMOCRACY.COM]. 324

Joseph S. Nye, Jr.,Technology.gov: Information Technology and Democratic Governance, in DEMOCRACY.COM supra , at 10-14 (reviewing opportunities and challenges of Information Age democracy). 325

Of course, the Digital Era challenge to civic republicanism based on deliberative decisionmaking by an engaged and well-informed citizenry goes well beyond the environmental realm.

103

5. Downsides of Cyberdemocracy More debate does not guarantee better policymaking. In fact, cyberdemocracy with its exposure to misinformatio n and dialogue among the uninformed could translate into a loss of deliberation and a step back for environmental decisionmaking. 326 More opinions being heard may lead to chaos and breakdown rather than higher quality decisions. 327 Hyperconnectivity — with a multitude of distractions only a click away — may translate into much less interest in deliberative policymaking. Even if some participants in the policy process stay focused, a flood of information could lead to decisionmaking on narrower (albeit deeper) basis with less of a focus on the broader context of a policy choice.328 In addition to the risk of distraction, the Information Age may empower some subsets of society and create a “digital divide” that leaves some citizens out of the decisiomaking loop. Already, those who are comfortable with computers find it much easier to participate in on-line discussions and to contribute comments to governmental decision processes that rely on Internet-based tools. Other observers fear that the Information Age will lead some people to talk only to those who are like-minded. Without real give-and-take, political debate is not possible. The extreme version of this fear can be seen in Sunstein’s caricature of the “Internaut” who gets all of his information from a highly

326

Dennis Thompson, James Madison on Cyberdemocracy, in DEMOCRACY.COM: GOVERNANCE IN A NETWORKED W ORLD (Elaine C. Kamarck & Joseph S. Nye, Jr., eds.) (1999) at 36-40 (noting that “directness” may be advised at the expense of deliberation). 327

Joseph S. Nye, Jr. Technology.gov: Information Technology and Democratic Governance, in DEMOCRACY.COM at 1014 (reviewing opportunities and challenges of Information Age democracy). 328

This crisis of civic republicanism, which depends on deliberative decisionmaking by an engaged and well-informed citizenry, goes well beyond the environmental realm.

104

tailored set of web sites producing a quotidian information sheet, which he mocks as the “Daily Me.”329 Others have expressed similar concerns about the challenges of cyberdemocracy.330

VIII. CONCLUSION Effective pollution control and natural resource management demands a great deal of information. If baseline data on harms, effects, and property rights are available, contractual exchange of environmental rights backed by a structure of contract and tort law may yield results that are generally efficient. 331 To the extent that information which parties need to negotiate is hard to come by or transaction costs are otherwise high, a range of other strategies for internalizing environmental externalities must be pursued. Identifying which institutional structures are best positioned to fill information gaps must therefore be understood as a central question in environmental law. Environmental decisionmaking through regulatory agencies may streamline the gapfilling process and reduce administrative costs. But this policymaking approach entails tradeoffs. In particular, overgeneralizations, which are inherent in the regulatory process, result in imprecise cost internalization and a de-emphasis on protecting environmental property rights. Information gaps thus loom large as a source of transaction costs — shaping our perceptions about the optimal means as well as the desirable ends of environmental policymaking. Information Age technologies promise to change the environmental protection calculus in fundamental ways. Computers, the Internet, advanced communications devices, sensors, and other 329

See CASS R. SUNSTEIN, REPUBLIC .COM (2002) (in the future, with the advent of television and Internet filtering technologies you will “need not come across topics and views that you have not sought out. Without any difficulty, you are able to see exactly what you want to see, no more and no less.”) However, Sunstein later backed away from this position in a response to a critique of Republic.com in the New York Review of Books, claiming that the “Daily Me” was only meant as a “thought experiment designed to cast light on the neglected requirements of a system of free expression,” and was in no way suggestive of “what most people are doing.” Response to James Fallow’s “He’s Got Mail,” NEW YORK REVIEW OF BOOKS , Mar. 14, 2002, available at http://www.nybooks.com/articles/15321. 330

See DEMOCRACY.COM, supra note 323. See also Rosa Tsagarousianou, Electronic Democracy and the Public Sphere: Opportunities and Challenges, in CYBERDEMOCRACY: TECHNOLOGY, CITIES , AND CIVIC NETWORKS 41-59 (1998), European Governance and Cyberdemocracy, in Governance in the European Union, a White Paper (European Commission, 2001), available at http://europa.eu.int/comm/governance/areas/group1/contribution_en.pdf. 331

Whether these results are just from a distributive perspective remains an issue.

105

innovations will make it easier and cheaper to address many existing sources of regulatory failure. Better data and analysis will mean that decisionmakers face fewer technical uncertainties. Strategic behavior will not disappear, but greater transparency and easy access to comparative data offer new ways to confront those whose positions emerge as outliers in environmental decision processes. Similarly, human cognitive limitations will remain an issue. But a more data -driven approach to environmental protection may provide some powerful new tools for unveiling irrationality. Special interests will continue to lobby decisionmakers in a self-interested fashion, but these same Information Age technologies will help to expose policy manipulations and limit public choice failures. Likewise, bureaucratic inefficiency or selfdealing will be easier to uncover and confront in a world richer in environmental performance measurement data and comparative analysis. These advances promise to shift the environmental policy Pareto frontier and to transform the optimal specificity of regulation. As a result, society’s mix of response strategies to environmental problems will evolve. Markets and contractual negotiations over outcomes will become more attractive. And better information makes it easier for regulators to harness market forces, which permits more tailored (thus more effective and efficient) interventions in the face of externalities. Even where traditional command and control regulation continues to be used, the Information Age seems likely to deliver gains in terms of refined governmental capacity to manage complexity, tolerate diversity, and ground environmental choices on firmer factual foundations. Lower cost information will allow us to re-think the ends as well as the means of environmental protection. As individual property rights become easier to delineate and protect, the cost of making corrective justice a policy goal goes down. To the extent that environmental problems arise from direct information failures—mistakes or waste—special opportunities to improve resource productivity and to identify and adopt “best practices” in both production and consumption are likely to emerge. Similarly, information disclosure strategies, such as eco-labels, may permit a shift toward greater emphasis on individual choice where externalities are not the central issue.

106

Some environmental problems cannot be resolved by the generation of more and better information. The technologies of the Digital Age may even aggravate some of the existing difficulties in environmental decisionmaking. Better information will not address questions of distributional equity. Nor will information fix human limitations with regard to risk perception. Although the precise contours of the path ahead are hard to predict, the Information Age creates the possibility of reduced information gaps leading to a restructured set of institutional arrangements and an environmental protection regime that is more refined, individualized, and efficient.

107