Interstate Compacts Can%t Solve Collective Bads Problems: The Case of LLRW
Dennis Coates
Department of Economics
University of Maryland - Baltimore County
(Coates@UMBC.edu)
Michael C. Munger
Department of Political Science
University of North Carolina-Chapel Hill
(Mike_Munger@unc.edu)
We thank Victoria Heid for extensive contributions to the research, and writing portions of the history sections, in this essay. We also thank Bill Shugart and Burton Abrams for their insightful comments at the meetings of the PublicChoice Society in Houston, Texas, April, 1996. Any errors are ours alone, however.
Abstract
A key tenet of the Republican %Contract with America" is the return of control of regulatory authority to the states. One policy area, the management and disposal of low-level radioactive waste, was identified as early as 1980 (in the LLRW Management Act) as a likely focus for gains from state cooperation and autonomy. These gains were to be achieved through %compacts," as set out in Article I, Section 10, § 3 of the U.S. Constitution. Compacts allow cooperation in managing public goods, such as watershed or fisheries management. We claim that compacts are singularly ill-suited, however, for managing public bads, particularly radioactive waste. In seeking disposal services, all states may want to join a compact ex ante. But each state has an incentive to defect from Compact agreements after that state is selected as %host" for a regional facility. We review legal literature that argues compacts are defective, and develop the logic of the defects in a simple cooperation game. Finally, we recommend some specific policy remedies for the disposal of public %bads" generally.
Interstate Compacts Can%t Solve Collective Bads Problems: The Case of LLRW
1. Introduction
The United States is nearing a crisis in its ability to manage the radioactive waste produced by nuclear power plants, medical processes and research, and a variety of industries. From the perspective of the nation as well as individual citizens and generators, the present state-based compact system for disposing of LLRW has failed. More than $400 million has been spent in site characterization, and extensive %studies" by geologists and public health experts.
Yet, in the more than 15 years that have passed since the %compact" system for low-level radioactive waste was established, not a single shovel full of dirt has been turned in building a new facility. Realistically, none are expected for years to come. Eleven interstate compacts have formed, but most of the Northeast states and Michigan remain unaffiliated with any compacts, and several are large producers of LLRW. In this paper, we give an argument for why this failure is the fault not of bad people, but of a bad system.
There are two types of radioactive waste: (1) High-level waste, mostly the remains of fissionable materials and highly radioactive reactor parts, as well as military waste. (2) Low-level radioactive waste (LLRW), which is mostly the byproducts of fission processes, or materials that have been contaminated by contact with radioactive substances. Current law (LLRWPA, 1980; amended 1985) encourages, but does not require, states to band together in regional %compacts" to provide for predictable, safe storage of LLRW.
Under the Low-Level Radioactive Waste Policy Act of 1980, Congress authorized interstate compacts to provide for regional LLRW disposal facilities. The LLRWPA allows interstate compactsto exclude waste from states not members of the compact. Without the LLRWPA, states could not exclude out-of-state LLRW because of two provisions of the U.S. Constitution: the Supremacy Clause and the Commerce Clause. Thus, the LLRWPA affords states a means of taking responsibility for waste produced within their borders, and sharing responsibility for waste produced in their region, without exposing themselves to the risk of serving as the dumping ground for the entire nation%s waste.
At least, that is the theory. We are not so sure. In earlier work (Coates and Munger 1992, 1995; Coates, Heid, and Munger 1992, 1994) we have analyzed specific problems in technology selection, host state identification, and coordination among states on LLRW policy. In the present paper, we extend our analysis to consider the constitutional impediments to solving collective %bads" problems generally.
A number of authors (Gordon, 1994; Hill and Weissert, 1995; McGinnis, 1994; Rabe, et al, 1994; Smith and Kearney, 1994; Vari, Cirincione, and Mumpower, 1994; Weiner, 1994; and Weissert and Hill, 1994) have considered constitutional problems and LLRW disposal before, of course. Our contribution is to provide a model of collective bads management that illuminates the incentive problems that make present policy generically unworkable. More simply, our thesis is that current policy has failed not because it has been poorly implemented, but rather because it is misconceived. Marginal reforms cannot work; we are going to have to start over.
2. The Good, The Bad, and the Radioactive
The nature of the services being contracted for is complex. Removal of dangerous waste beyond the risk of contamination of human populations is a %good," in the sense that people want it evenif it has a positive price. The waste itself, of course, is a %bad," an undesirable by-product of generating electricity in nuclear power plants, or of producing pharmaceuticals or smoke detectors, or performing medical diagnostic procedures.
It is often useful simply to treat bads as the mirror image of goods (Hahn and Hester, 1989). But as Ostrom (1990) points out, it is also useful to consider the unique context, and incentives, disposing of public %bads" provides. In particular, the specification of property rights, and liability, become key issues. As we shall see, the specification of rights and liabilities to the public %bad" of LLRW is unclear; the %compact" system has, if anything, made the problem worse.
The LLRW disposal problem can be summarized as achieving a public good for the nation (safe and predictable isolation of dangerous radioactive waste from human population), but as a local public bad: People in the host state must live near a disposal facility. Prochaska (1986) summarizes the conflict clearly:
The goals of the compact system clash with the motivations of the individual states. The compact system requires that the states negotiate with each other to voluntarily create regional compacts. Yet the forty-seven state that do not have disposal sites within their borders do not want them, and they are content to let the present situation to persist as long as possible . . .
In contrast, the federal government wants a stable, quick, and politically palatable solution to the low-level waste disposal problem....The states had the same broad goals as the federal government, but when the states pursued these goals within the compact system, the intended results were not reached. A state, first and foremost, wants to ship it radioactive waste elsewhere. (p. 401, emphasis added).
3. The Constitutional and Historical Background
3.1 History: Compacts in LLRW Disposal
The most recent major uses of interstate compacts have both involved waste disposal: The Solid Waste Act (1976), and the Low Level Radioactive Waste Policy Act (1980). Unlike the compacting provisions in the Solid Waste Act (in which the states apparently have had no interest), the compact approach embodied in the LLRWPA of 1980 was actively sought by the states.
It is interesting, and instructive to compare the "success" (which we will argue is chimerical) of the LLRW compact system with the complete inaction in solid waste compacting. Cain (1983) attributes the difference to a comparison of the nature of the waste streams. He writes:
Radioactive wastes are generated in small volumes, and they are expensive to manage because they demand sites with high development and licensing costs. Transportation costs, however, are small because of the small volume of material involved. Thus, the total waste stream can be handled at a few sites located throughout the country. Hazardous chemical wastes, on the other hand, are generated in volumes several orders of magnitude greater than those which must be handled in the low-level radioactive waste system, and large quantities are generated in each state. Consequently, with the exception of a few low-volume, extremely hazardous materials, most of the economic factors which make regional management attractive for radioactive wastes do not apply to the hazardous chemical waste problem. (p.783).
With the benefit of hindsight, it would appear that the early optimism aroused by the formation of interstate compacts for LLRW was misplaced, and not just because no new LLRW disposal facilities have been built. The promise of %success" of the LLRW compact system has not been realized. We argue that both types of compacts, those to manage solid waste (none of which have formed) and those to manage LLRW (more than 15 of which have formed at one time or another), share the same fundamental, and irremediable, flaw. Compacts for managing public goods afford states a means of taking responsibility and realizing gains from cooperation. Compacts for managing wasted disposal are primarily mechanisms to avoid responsibility within the federal system.
A Brief History
Not until after WWII did Congress begin discussing peaceful uses of atomic energy, but even then little thought was devoted to radioactive waste disposal. In 1946, after much debate over whether the military should retain control, Congress passed the first Atomic Energy Act giving responsibility for the regulation of all aspects of atomic energy to the Atomic Energy Commission (AEC). The Act placed the AEC under civilian control but left "military applications" of atomic energy to a Military Liaison Committee with direct access to the President (Mazuzan, 1985, p.3). Only gradually did AEC and Congress contemplate civilian uses of the atomic energy.
After resigning as Chair of the AEC, David E. Lilienthal published an article in 1950 describing his frustration in trying to open the government monopoly on atomic energy. He wrote, "[N]o Soviet industrial monopoly is more completely owned by the state than is the industrial atom in free-enterprise America." (Mazuzan, 1985. p.18). Lilienthal's attack launched a process to allow the use of atomic technology by private enterprise, formally granted through the 1954 Atomic Energy Act.
Tight Federal control of all aspects of atomic energy did not lend itself to public accountability over radioactive waste disposal. Details were kept confidential for %national security." Even as atomic energy began to be used for peaceful purposes by the private sector, issues such as reprocessing of spent fuel rods remained within the purview of national security because of concerns that terrorists might gain control of weapons-grade radioactive materials. This tradition of secrecy, combined with the invisibility of radioactive pollution and debate over its long-term health effects, has made accountability for safe disposal extremely difficult from the start.
The AEC disposed of most solid low- and intermediate-level radioactive waste through shallow land burial at Federal installations. Drums of solid waste were also dumped in the Atlantic and Pacific Oceans. For low- and intermediate-level liquid waste, the AEC continued traditional methods of "dilute and disperse." Most liquid wastes were treated and then released into waterways; others were packaged in drums for land or ocean burial.
Ongoing weapons testing in the 1950s by the U.S. and other countries sharply raised fallout levels in the atmosphere, particularly in 1959. As more information appeared about the levels of strontium 90 in the atmosphere and in wheat and milk samples from bomb test fallout, the AEC was perceived to be down-playing the dangers prompting public concern about the AEC's performance in safeguarding public health against radioactive contamination. Public concern focussed on atomic weapons testing, but growing distrust of the AEC spilled over to the Commission's assessments of risk in general. Then, as now, the public made little distinction between types of radioactive waste. As fallout levels in the atmosphere dropped after 1959, so did public concern. Even so, the public had learned to distrust the AEC's risk analyses. Subsequent efforts to site radioactive waste disposal facilities or license storage of LLRW in New England met with citizen opposition.
State governments first become involved in atomic energy during this period with passage of the Atomic Energy Act Amendments of 1959. Those Amendments allowed states who wished to administer their own radiation protection programs toapply for "Agreement State" status. Agreement State programs could not be less stringent than Federal AEC regulations. In 1962 the AEC began taking license applications from private companies to operate LLRW disposal facilities. These new sites were to be owned by the states and operated by commercial firms under Federal regulatory guidelines (or Agreement State regulatory programs if more stringent). By the early 1970s six facilities were operating throughout the U.S., all in Agreement States. Facilities operated in Maxey Flats, KY, West Valley, NY, Sheffield, IL, Beatty, NV, Richland, WA and Barnwell, SC.
By 1979, however, only three facilities remained open - in Washington, Nevada and South Carolina. The Maxey Flats and West Valley facilities were shut down due to operational and water management problems on the part of the civilian contractor. The Sheffield facility had reached storage capacity and the Nuclear Regulatory Commission (NRC, formerly the AEC) denied US Ecology's request to expand.
The closure of the three commercial facilities meant increased volumes for the remaining three facilities. The governors of the three sited states felt their states were bearing an unfair share of the burden of the nation's LLRW disposal. They could not refuse out-of-state LLRW, however, without violating the Interstate Commerce Clause. To force a comprehensive reassessment of the national policy on radioactive waste management, the governors of Washington and Nevada temporarily closed their state facilities in 1979 for minor packaging and transportation violations.
The Low-Level Radioactive Policy Act of 1980 (U.S. Public Law 96-573 (42 USC 2021d))
Congress responded to the concerns of South Carolina, Nevada and Washington by establishing a state-based system of siting LLRW disposal facilities through the Low-Level Radioactive Waste Policy Act of 1980 (LLRWPA). The LLRWPA was all that emerged in the closing days of the Second Session of the 96th Congress after what had begun as an effort to enact a comprehensiveNuclear Waste Policy Act. The low-level radioactive waste (LLRW) provisions of the comprehensive act were spearheaded by the South Carolina delegation and became a bill in their own right by default when agreement on the more comprehensive radioactive waste policy proposals proved impossible. Virtually no disagreement appeared over the LLRW provisions.
Sentiment in Congress to address the radioactive waste management issue had been growing for several years. Senator Charles Mathias (R, MD) had first attempted (and failed) to find a solution with the Nuclear Waste Management Act of 1977. In that session, the newly created Department of Energy was given authority for developing a nuclear waste repository; almost immediately after its creation, however, DOE moved back the target dates for construction of a new facility so little progress was made. Again in 1978, Senator Charles Percy (R, IL) introduced a Nuclear Waste Management Act, and several committees held hearings on nuclear waste in July of that year, though again no agreement was forthcoming. And in 1979, during the First Session of the 96th Congress, Senator Mathias introduced the Nuclear Waste Management Reorganization Act. But that session would prove no more fruitful than the last in resolving radioactive waste issues.
In Fall, 1979 the Natural Resources Defense Council published an article (Lash, 1979) which attributed the moving target dates and lack of substantial progress on the waste disposal issue to "gross technological optimism and neglect of non-technological issues." USDOE's overly optimistic predictions for opening of a repository, repeatedly followed by delays and target date revisions, cast doubt that any solution would be possible in the near future. In response, states began placing moratoria on nuclear plant licensing until the radioactive waste problem could be resolved. California was the first to pass such legislation (in 1978); a host of states followed. The sentiment behind this flurry of legislation was voiced by Senator Gary Hart (D, CO), chair of the subcommittee on nuclear regulation: "If the word 'scandal' can be attached to nuclear power, it is that this industry has been permitted to expand for two and ahalf decades without an acceptable solution for [radioactive] waste disposal."
On July 10, 1979, the Governors of the three sited states, Washington, Nevada and South Carolina, sent a letter to the Federal Government charging that the NRC and the Dept. of Transportation had shown "serious and repeated disregard" of existing regulations controlling the shipments of nuclear waste. Governors List, Ray and Riley related in their letter "the total lack of corrective measures" taken by NRC and Dept. of Transportation to reduce the hazards in shipment of radioactive waste. Some have suggested these transportation and packaging violation charges were only minor and were used by the sited states to force Congress to reassess the LLRW disposal policy.
But Senator Metzenbaum cited the Governors' letter as evidence that the NRC and Dept. of Transportation should be forced to greater accountability -- hence his proposed amendment in 1979 to require the NRC to inform states of radioactive waste shipments. Senator Metzenbaum accepted the sited states' Governors' claims, and argued that the violations were not simply a trivial and convenient misstep by the NRC which the Governors were exploiting to curtail LLRW from outside their state borders. Metzenbaum claimed that by requiring the NRC to notify states of shipments in advance, states could better inspect the shipments and perhaps route them during off-peak travel hours. The Governor's tactic was effective: a few weeks after the letter, Nevada and Washington closed their facilities temporarily.
In Fall of 1979, Governor Dixy Lee Ray of Washington reopened the Hanford facility to LLRW on the condition that USDOE and NRC officials, together with Congress, attempt a regional solution to the LLRW problem. The Nevada facility also reopened for limited amounts of waste. The two states had made their point: In the face of heightened public concern and state desires to avoid siting of new radioactive waste disposal facilities within their boundaries, the Second Session of the 96th Congress seemed more promising for resolving the radioactive wastemanagement issue. S. 2189, which began its tortured life as the Nuclear Waste Policy Act, was designed to establish a program for Federal storage of spent fuel from civilian nuclear power plants and a Federal policy and program for the disposal of nuclear waste from civilian activities. It was intended as a comprehensive solution to civilian radioactive waste disposal. Several amendments were added, including the Percy-Glenn amendment which would have provided for "consultation and concurrence" between States and the Federal Government on where disposal facilities might be located; a Hatfield amendment regarding full-cost recovery; a Hart-Simpson amendment regarding away-from-reactor (AFR) storage; and a Thurmond-Hollings amendment, regarding LLRW disposal. The Thurmond-Hollings amendment was virtually all that would survive.
Similar to bills introduced in the previous session, the South Carolina delegation's amendment made each State responsible for the disposal of non-Federal radioactive waste generated within its borders. It asserted that LLRW can be most safely and efficiently managed on a regional basis. To encourage regional disposal the amendment authorized States to enter interstate compacts with the power to exclude out-of-compact waste from a compact's disposal facilities without being in violation of the Interstate Commerce Clause. Also, USDOE was authorized to provide technical and financial assistance to the states for this purpose. These three provisions -- for state responsibility, compacting and exclusionary authority, and USDOE assistance -- were the same recommendations made to President Carter by "The State Planning Council on Radioactive Waste Management" of which Governor Riley of South Carolina served as Chair. According to Senator Strom Thurmond's arguments on the floor on July 29, 1980, the State Planning Council had unanimously adopted the following resolution:
The national policy of the U.S. on LLRW shall be that every State is responsible for the disposal of LLRW generated by non-defense related activities within its boundaries and that States are authorized toenter into interstate compacts, as necessary, for the purpose of carrying it this responsibility. ( Congressional Record , 19 May 1980)
3.2. Constitutional Issues
There are three portions of the U.S. Constitution directly relevant to our discussion: the supremacy clause , the commerce clause , and the compact clause , which describes agreements between states and other entities. Each of these three clauses is reproduced below.
% Supremacy Clause (Article VI, clause 2):
This Constitution and the laws of the United States which shall be made in pursuance thereof and all treaties made or which shall be made, under the authority of the United States, shall be the supreme law of the land, and the judges in every State shall be bound thereby, anything in the Constitution or laws of any State to the contrary notwithstanding.
% Commerce Clause (Article I, section 8, clause 3):
The Congress shall have power: ... To regulate commerce with foreign nations, and among the several states, and with the Indian tribes.
% Compacts among States (Article I, section 10, clause 3):
No State shall, without the consent of Congress . . . enter into agreement or compact with another State, or with a foreign power . . .
These three portions of the Constitution have been interpreted to imply that federal law is preeminent over state law within the federal jurisdiction, that states have only limited power to regulate interstate commerce, and that states may only enter into agreements with other states with the explicit, or directly implied, consent of Congress.
Interpretations of the Supremacy and Commerce Clauses
The Commerce Clause of the U.S. Constitution clearly gives Congress the power to regulate commerce among states. The U.S. Supreme Court has found that this clause implies a nearly exclusive power to regulate interstate commerce, with the states retaining only limited ability to regulate private enterprise that crosses state borders, in spite of the fact that the clause itself does not clearly delineate the boundaries between Congressional and state powers.
In Gibbons v. Ogden, the Court gave the first real definition of "interstate commerce." Writing for the majority, Justice Marshall appeared to consider, but did not fully adopt, the notion that the federal power is exclusive. The power to inspect, or enforce other state laws, derived (in Marshall's view) from the states' police power, whereas the power to regulate interstate commerce belonged to the Congress. In instances where state and federal law conflict, the federal law prevails, as state laws must be subordinated because of the supremacy clause.
After Cooley v. Board of Wardens, the power to regulate commerce was recognized as %concurrent," rather than exclusive. No obvious or authoritative doctrine for determining the extent of the power was available, however, until Southern Pacific Co. v. Arizona. The doctrine derived from this case is a balancing of the legitimate needs to ensure citizens' safety within the state, and the burden and costs placed on commerce among the states. In effect, the doctrine held regulations that are reasonable for ensuring safety are acceptable; regulations that discriminate against interstate commerce (either intentionally or in effect) are not.
Consequently, economic protection of the local market from out-of-state competition is definitely not acceptable, unless the operation of the %trade" is manifestly unsafe or likely to cause physical harm to citizens. This turns out to be very important in considering multi-state management of public bads, because local %protection" (of real estate values, or attractiveness to new business) generally cannot be used as a rationale for regulation by a state, even on traffic in waste within its own borders. This doctrine, however,is uncomfortably applied to management of bads, to say the least. In what follows, we will assume that the interpretation of the %commerce clause" is constant, though we note that it is possible that the doctrine we have discussed could evolve in response to pressure from the states.
Compacts
An %interstate compact" is a binding agreement among states, approved by each party state%s legislature, and by Congress. The first uses of interstate agreements predate the U.S. Constitution, of course. Compacts in the technical sense we will use the term derive from the inclusion in the Constitution of the clause already quoted above: "No State shall, without the consent of Congress... enter into agreement or compact with another State, or with a foreign power..."
The Constitution defines neither "agreement" nor "compact;" later court interpretations of this vague provision have tended to interpret the two words as having different meanings:
If either or both words had possessed precise meanings at common law when the Constitution was framed, we might assume--as with other provisions of the Constitution--that the prior meanings were the ones intended by the founding fathers. Research has shown that both terms were in use before 1789 but that the usage was fuzzy. Agreement and compact appear to have been used as synonyms quite frequently, but on other occasions nice differences may have been intended. (Zimmermann and Wendell, 1951, pp. 30-31).
A general, but not completely accurate, summary would call "agreement" the generic term used to describe arrangements, or even formal contracts, among states, and "compact" the specific term used with reference to more formal and binding agreements. In fact, "It is sometimes said that an interstate compact is a treaty between states." (Zimmermann and Wendell, 1951, p. 31).
If anything, the notion of a treaty does not go quite far enough to describe the formality of a compact arrangement, since an individual state legislature is bound by the compact: "a state statute which conflicts with an interstate compact is invalid and unenforceable." It is this handing over of sovereignty, within the federal system, by a group of states to a collective authority or commission that is the key constitutional test: If sovereignty is given up, the U.S. Congress must approve the compact arrangement. Otherwise, the contract can be treated as a more informal agreement.
The two key cases that established this test were Holmes v. Jennison and Virginia v. Tennessee. These two cases together suggest the following test for when Congress must take affirmative action (i.e., pass legislation) to validate an agreement (thereby making it a "compact"), and when the Congress can implicitly accept the validity of an agreement simply by doing nothing. The test is this: Congressional consent is not explicitly required on matters that have no material effect on the United States or the relative political strengths of the states and federal government. The Congress must give its affirmation, however, if there is some agreement "tending to increase the political power of the States, which may encroach upon or interfere with the supremacy of the United States." (Hay and Rotunda, 1982, p. 187).
Congress has authorized interstate compacts to form for matters including airport development, fishery and watershed management, crime control, and the management of hazardous wastes. No distinction has been made between the type of good or activity being managed. We claim, however, that one distinction (between collective %goods" and %bads") turns out to be crucial in predicting the success or failure of the compact itself.
Compacts Afford a Constitutional Mechanism to Secure Goods, and Avoid Bads
Compacts are best suited for only certain types of interstate problems; specifically, compacts work best in management of publicgoods, secured through collective action.. They may be an ideal vehicle for building joint administrative machinery or settling narrow disputes such as disagreements over boundaries. The advantage of the compact is that its actions, or agreements it arranges, are beyond the power of state legislatures to modify or veto. Because the agreements themselves internalize within the states in the compact the gains from cooperation, the ability to cooperate in once-and-for-all settlements is enhanced.
But, as Zimmermann and Wendell argue,
"The interstate compact...has not proved a satisfactory medium for continuous and complex planning and development. 'Administration resulting from compact provisions is almost sure to be unbusinesslike because it will lack independence, initiative, flexibility, and coercive authority.' Moreover, compacts have tended to encourage 'particularism, selfishness, and jealousy on the part of individual states, rather than the broad regional cooperation which is to be desired.' Some other method of governmental action may be possible and preferable, which would emphasize regional interests rather than those resulting from artificially drawn state lines" (p.110-111).
Compacts are needed in situations where states cannot cooperate intensively or continuously enough, or where states are too small or too porous to be appropriate units in a regional structure In the case of LLRW disposal, this problem is clear: Creating 50 state LLRW disposal facilities would be ludicrously redundant -- making some sort of cooperative effort necessary. (Coates, Heid, and Munger, 1994, estimate that a maximum of five facilities are even viable in the U.S.) But achieving such gains from cooperation has been blocked by just the sort of forces Zimmermann and Wendell discuss: particularism, selfishness, and the artificiality of states as decision-making units.
Zimmermann and Wendell did not use the NIMBY (%not in my backyard") acronym, of course. Still, their description ofproblems for compacts clearly implies that NIMBY forces within states will make compacts ill-suited as a mechanism to realize cooperative gains in waste disposal. In the case of LLRW disposal, regional solutions can improve economic efficiency, minimize the number of communities who must grapple with siting a controversial facility, and identify the most geologically suitable (and therefore safest) sites -- all without the constraints imposed by state boundaries. But the logic of regional gains can be defeated by within-state NIMBY pressures, against which compacts are virtually powerless.
Development of an interstate waterway system, protecting a common pool resource such as a fishery, or choosing a regional airport cargo site, all involve the provision of a public "good." Each state might well want to be the host for a regional air cargo facility, for example, so that the state actually chosen by a compact among the states is happy to cooperate in the site selection activities.
Siting a LLRW disposal facility is a "bad," however, for the state, and avoiding the %host" rule is a goal shared by all the states in a compact. Consequently, incentives for compliance and cooperation are actually reversed. If a disposal facility must be hosted at all within a state, officials of that state will always feel the political pressure to drop out of the compact and seek smaller LLRW disposal facilities over larger facilities. The fact that this means more LLRW disposal facilities nationwide has no effect on the incentives facing host state officials.
As Prochaska (1986) notes, with hindsight we can easily see these incentive problems at work. First, %The states did not form regional compacts as quickly as sponsors of the LLRWPA expected, nor did the states group themselves into the anticipated six to eight large regional compacts." (p. 386). Second, the initial failure shows little sign of fixing itself, because the incentive problems are generic:
To form a compact, the party states must consent as sovereign powers. A state therefore will join a compact only if its terms are suitable and advantageous; no state willwillingly join a compact detrimental to its interests. A state as a sovereign will also feel very little outside pressure, and that only obliquely, to act quickly on a compact. ... (Prochaska, 1986, p. 389).
Lawmakers recognized some of the problems, and built into the 1985 amendments a provision designed to force states to take responsibility for waste produced within their borders. Under the LLRWPAA of 1985, states that did not make arrangements for LLRW disposal by 1996 would be required to take title to and possession of LLRW at the request of any generator within their borders. The "take title" provision was the ultimate incentive for states to compact with each other for LLRW disposal.
In the fall of 1991, however, New York state challenged the LLRWPAA and its "take title" provision on the grounds that the federal government had exceeded its authority and the law "violated the principles of state sovereignty and federalism guaranteed by the 10th Amendment to the Constitution." Rather than being simply an incentive, New York State argued that the "take title" provision commanded the states to take ownership of LLRW if they did not adhere to federal regulation. In its decision June 19, 1992, the Supreme Court upheld the LLRWPAA but found the "take title" provision unconstitutional. Justice Sandra Day O'Connor wrote the majority decision, and stated flatly: "Where a federal interest is strong enough to cause Congress to legislate, it must do so directly. It may not conscript state governments as its agents."
The "take title" provision was the enforcement mechanism of the LLRWPAA. It gave states a strong incentive to make LLRW disposal arrangements for their in-state generators. Without strong incentives for states to accept responsibility for ensuring for safe and predictable waste disposal, the compact system was doomed.
So far, we have only argued in generalities, and relied on other sources for analysis. In the next section, we present a simple game-theoretic model of %host" state selection, and stateperformance under compact agreements. We will use this model to argue that the failure of the compact system is unavoidable, and generic to problems of solving coordination dilemmas in the management of public bads.
4. A Model of Host State Identification
In this section we develop a simple noncooperative game, designed to capture the incentives facing each state in a compact for the provision of some public service. The size of the compact is taken to be fixed but arbitrary, with at least two states. To abstract away from complexity without strategic interest, we have only two entities: state %i" and all other states %o". There are n states total, so n-1 states comprise the other states. We will specify a payoff function for each entity i and o in terms of monetary benefits associated with the existence and possible hosting of the facility. Additionally, this game structure we specify is common knowledge to all the states.
Each state has to choose between volunteering to host the facility and not volunteering. If no state volunteers, a host is chosen at random. To be able to make precise statements about the variables that affect these decisions from the perspective of a %rational" state (taking states as unitary actors), it is useful to introduce some notation.
NOTATION:
Let Mj = benefits to state j from the existence of a publicly provided facility
Mj* = benefits to state j from being host to the facility; in the case of a noxious facility, this variable may be negative
Pj = state j's prorated share of the cost of development and operation of the facility
If state j hosts the facility, its payoff is:
uj =Mj+Mj*-Pj
If state j does not host the facility, but one is provided, the payoff to j is:
uj= Mj-Pj
If no facility is provided, payoff is zero.
At the point of decision, each state knows that it is possible that more than one state volunteers or that no state volunteers. In either case, the host is selected at random. Hence, the expected payoffs for either of these two situations must be determined. The expected payoffs to each state are the same whether all or none volunteer. Without loss of generality, therefore, consider the case where both players volunteer. The expected payoff to i is:
(Mi+Mi*-Pi )/n + (n-1)(Mi-Pi)/n = Mi-Pi +Mi*/n
For o, the payoff is:
(n-1)(Mo+Mo*-Po )/n + (Mo-Po)/n = Mo-Po +(n-1)Mo*/n
Suppose the facility is a public good. That is, every state derives positive net benefits from hosting it; Mj+Mj*-Pj > 0. In this case, it is easy to show that volunteering is a dominant strategy. Consequently, the equilibrium to this game is for each state to volunteer, setting in motion the random selection of a host.
After the random selection, the incentives of states may change. For example, suppose that state i is not chosen as host and Mi-Pi<0. In this case, i wants to block provision of the facility, even though i volunteered to host it. This outcome may be that of the superconducting supercollider the construction of which has been halted in Texas. Many states volunteered, indeed, actively lobbied, to host this Federal research facility. Now, however, representatives of many of those same volunteer states have pushed for putting an end to its funding.
Of course, Mi-Pi may be greater than zero for all states. In this case, provision of the facility will proceed as planned for each state recieves a positive payoff regardless of wich is chosen as host. One might think of successful interstate compacts for watershed management, regional transportation authorities, and others as examples of this situation.
But suppose instead that Mi+Mi*-Pi is negative because Mi* is negative. That is, suppose that serving as the host to the facility carries costs to the state beyond the prorated share of devemopment and operation. Such might be the case for hosting a noxious facility such as a radioactive waste facility, a hazardous waste incinerator, or a prison. Now the dominant strategy is not volunteer. And when no state volunteers, a host is selected at random.
The randomly selected host receives a payoff of Mi+Mi*-Pi < 0 if it acquiesces in its selection. It can get a payoff of zero bydelaying development of the facility. Moreover, each of the non-host states breathes a sigh of relief and presses for the selected state to provide the facility as soon as possible.
Consider, for example, the case of the Midwest Compact. Michigan was part of the Midwest Compact and had been selected as the host state in 1987. The state government had conducted a statewide survey and narrowed the possible sites down to 3 areas. Site characterization these 3 sites continued, until May 1990 when all three were eliminated as unacceptable under state regulations. Michigan's intent was to go back and begin further study of the remaining 75 sites which had made the cut after the initial statewide survey. They wanted to do a little more review of all of them to see if any would meet state criteria.
But at this point the Michigan site selection process cme under fire from the Midwest compact. The compact claimed that Michigan was simply stalling, had no real intent to site a facility., and that the criteria were designed to be too difficult to meet. In summer 1990 the Midwest Compact stopped financial support and peremptorily expelled Michigan from the compact.
A new Governor (Engler) was elected in November 1990. He was opposed to siting a facility in Michigan, because federal law would result in far too many LLRW disposal sites (for evidence that there is some basis for this claim, see Coates, Heid, and Munger, 1994).
The remaining members of the Midwest Compact (sans Michigan) have reformed to resume siting their own LLRW facility. Ohio, originally selected as second in host state rotation, now moved into host state position for the Midwest Compact. But for all practical purposes, the siting process is also stalled in Ohio. Before the siting process can begin again, Ohio must adopt its own state-specific enabling legislation to establish criteria for choosing a site, identify the responsible state agencies, create a schedule, and possibly select the LLRW disposal technology and incentives for potential host communities. Like Michigan in 1987, however, Ohio has raised questions about compact provisions on shared liability andcosts among party states. Ohio does not expect to have a LLRW facility open before 2000, and has made no real progress toward even selecting a site.
A further example is the experience of North Carolina in the Southeast Compact. North Carolina was chosen as %host state" in 1986, using criteria developed by the Compact. The state conducted a statewide search for suitable areas, and narrowed the process down to four finalists in 1992. In 1993, two sites (one in Richmond County, on the South Carolina border, and one in Wake County, near Raleigh) were identified as the best candidates for licensing and operating a LLRW disposal facility.
Eventually, in 1993 the Wake County site was selected as the preferred site, with construction expected to begin in 1995. Detailed characterization and field studies were begun. The result has been more than $90 million in payments to consultants, attorneys, and geologists. Nearly four years later, however, no construction has been accomplished, and the license application is languishing in the state Department of Radiation Protection. There is little prospect of the license being granted anytime soon.
South Carolina, home of the operating facility for Southeast Compact waste, reacted angrily. Claiming that North Carolina was dragging its feet, South Carolina first threatened to close its facility completely. Later, in the summer of 1995, South Carolina decided to leave the Southeast Compact and accept LLRW from the entire nation, except that they would exclude waste from North Carolina generators.
In the spring of 1996, South Carolina appears ready to stay open indefinitely, North Carolina is prepareing for a new round of studies (and payments to consultants), and the Southeast Compact is considering a move toward designating Virginia as the new host state.
In both the Midwest Compact, and Southeast Compact, we have observed the same pattern: the Compact solicited volunteers to host the facility, and when none were forthcoming tried to impose %host" status on a member state. When that state balked, long andexpensive delays ensued, after which another state was selected (the Midwest switched from Michigan to Ohio) or another round of studies was commissioned (North Carolina is preparing to start over in its search for geologic information).
We suggest that this pattern is in fact generic, and unavoidable: when the payoffs to the host are les than those of going it alone, any state designated host will resist. If this pattern of preferences is shared by all states in the compact, the compact will fail to site a facility. The reason, it turns out, is that the compact system does not provide sufficient institutional mechanisms to solve the collective action problem, or to enforce compact decisions. Compacts in such situations, therefore, are doomed to fail.
One might object that in the Southwestern (California volunteered), Appalachian (Pennsylvania volunteered), Central Midwest (Illinois volunteered), Northwestern (Washington volunteered), and Texas (Texas volunteered) Compacts individual states have volunteered. All this means is that there are settings in which Mi+Mi*-Pi, the payoff from hosting the facility, is positive. Except for the Northwestern Compact, each of these compacts involves one very large generator hosting a facility with one or more very small producers along for the ride. The Northwestern Compact is also unique in that it had an operating facility in the volunteer host, hardly a scenario which the compact system should boast of as a success. On the other hand, none of these non-sited compacts has opened, or even licensed, a facility yet.
Finally, we must contrast the situtation with respect to llrw policy with the siting within states of a regional solid waste landfill, prisons or other public bads. This is necessary because nothing in the formal model rules out application of the model to these latter issues. American political and governmental institutions do distinguish these situations. The states in the United States exist as independent entities in the American federal system. The U. S. Constitution explicitly recognizes the existence of states in several ways. Indeed, individual states were required to ratify thedocument, and it did not take effect until a minimum number of states had consented.
The compacting clause serves to maintain the balance of power between the federal government and the states. States must get federal permission to engage in any joint activity which would tend to shift that balance of power. Moreover, states are actors in the llrw policy debate because under the LLRWPA and its amendments it is the states that must conduct the site selection process. And the states determine the criterion under which siting will occur, subject only to licensability by the NRC.
In the case of regional landfills and prisons, only one level of government is involved. States must selesct the site and pay the costs. Local governments can raise objections, but they are powerless to block the actions decided upon by the state because, ultimately, they exist at the whim of the state. Moeover, local communities cannot agree to host facilities without the consent of the state. In fact, some communities have volunteered to host llrw faciities but were not able to when the rest of their states objected.
Our model works, therefore, for the federal llrw policy precisely because states are real players in the game. It does not work for the siting of state prisons or other local noxious facilities because counties and cities are not real players.
5. Conclusion
The interstate compact system ratified by the Low-Level Waste Policy Amendments Act of 1985 may be unable to handle the waste disposal problem. .. [T]he compacts themselves may not hold together over time. Instead, the regional compacts could fragment into smaller groups of one, two, or three states. This fragmentation could, in turn, cause the price of waste disposal to increase dramatically, increase the environmental danger from low-level wastedisposal, and result in eventual abolition of the compact system entirely. (Prochaska, 1986, pp. 383-4).
How prophetic the words of Prochaska seem now. We have shown that the experiences of the compact system for low level waste management are consistent with a simple game. From this game we observe that the difficulties of the compact system for LLRW management have not arisen from incompetence or bad motives of any of the participants. Rather, the problems of the compact system are the product of inherent contradictions in the use of compacts to manage public bads. Tinkering with the system won%t help; we must start over again from scratch.
However, as we have argued in other settings (Coates, Heid, and Munger, 1992, 1994), the problem is complex. The present system is too centralized, and too decentralized, to work effectively. There is too much focus on states as players in a veto game, with too little focus on communities and the compensation to the citizens who will be directly affected. Indeed, even in the simple game we develop here, if the host state compensation is raised, and the payments of no-host states raised accordingly, it may be possible to induce host states to acquiesce to developing a facility without delay. That South Carolina has reopened the Barnwell facility, and Washington has kept the Hanford site in operation suggests that states can derive benefits in excess of costs from host state status. On the other hand, states cannot coordinate effectively (because of the Commerce Clause in the U.S. Constitution) to avoid developing redundant, or even financially inviable, facilities.
The solution is for the U.S. Congress to establish some number (perhaps four, perhaps five) of zones or regions, with approximately equal volumes of LLRW production. These regions can then be given the charge to find a community that will accept the waste for a certain package of incentives. It is also clear that the incentive package must include payments to the state. If no community is willing to accept the %reverse auction" bid (higher and higher compensation for accepting the local bad), the regionalauthority will be charged with contracting with another region that has sited a facility.
A more radical solution would be pure privatization of the disposal function, with private firms accepting and disposing of the waste from all generators. The widespread perception that the negative externalities associated with living even in the same county as a LLRW disposal facility make the pure privatization option unlikely to survive tests of political feasibility.
References
Burns, Michael E., editor. (1988). Low-Level Radioactive Waste Regulation: Science, Politics and Fear . Chapter 3, "Interstate Compacts for Low-Level Radioactive Waste Disposal: A Mechanism for Excluding Out-of-State Waste," by Robert L. Glicksman. Lewis Publishers, Inc., Chelsea, MI.
Cain, Jonathan T. (1983). "Hazardous Waste Imports," in Natural Resources Journal . Vol. 23, October 1983.
Coates, Dennis, and Michael Munger. "Guessing and Choosing: A Multicriterion Decision on a Disposal Technology for Low Level Radioactive Waste." Journal of Public Policy August (1992).
_______. "Strategizing in Small Group Decision Making: Host State Identification in the Southeast Compact" Public Choice . 82 (1995): 1-16
Coates, Dennis, Victoria Heid, and Michael Munger. The Disposal of Low-Level Radioactive Waste in America: Gridlock in the States . St. Louis, MO: Center for the Study of American Business, Occasional Paper No. 119 (December 1992).
_______. "Not Equitable, Not Efficient: U.S. Policy on Low-Level Radioactive Waste Disposal." Journal of Policy Analysis and Management . 13 (1994): 526-541.
Hahn, R.W., and G. L. Hester. 1989. %Marketable Permits: Lessons for Theory and Practice." Ecology Law Quarterly . 16: 361-406.
Hill, Jeffrey S., and Carol S. Weissert. (1995). %Implementation and the irony of delegation: the politics of low-level radioactive waste disposal." Journal of Politics . 57: 344-370. Low-Level Radioactive Waste Policy Act, Amendments, Etc. (1986). Public Law 99-240, January 15, 1986. Washington, DC: U.S.G.P.O.
Mandelker, Daniel R., and D.C. Netsch and P.W. Salsich. Jr. (1977). State and Local Government in a Federal System , 2nd ed. The Michie Company, Law Publishers, Charlottesville, VA.
McGinnis, Michael Vincent. 1994. %Collective bads: the case of low-level radioactive waste compacts." Natural Resources Journal . 34:563-588.
Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action . New York: Cambridge University Press.
Rabe, Barry G., William Gunderson, Hilary Frazer and John M. Gillroy. 1994. %NIMBY and maybe: conflict and cooperation in the siting of low-level radioactive waste disposal facilities in the United States and Canada." Environmental Law . 24: 67-122.
Smith, Ande A., and Richard C. Kearney. 1994. %The low-level radioactive waste siting process in Connecticut: anatomy of a failure." Policy Studies Journal . 22: 617-31.
Southeast Compact Commission. (1987). Low-Level Radioactive Waste Management in the Southeast: A Primer .
Vari, Anna, Patricia Reagan-Cirincione, and Jeryl Mumpower. 1994. LLRW Disposal Facility Siting: Successes and Failures in Six Countries . Boston: Kluwer Academic Publishers.
Weiner, Richard D. 1994. %Federalism and the disposal of low-level radioactive waste." Natural Resources Journal . 34: 197-223.
Weissert, Carol S., and Jeffrey S. Hill. 1994. %Low-level radioactive waste compacts: lessons learned from theory and practice.." Publius . 24: 27-44.
Zimmermann, F., and M. Wendell, (1951). Interstate Compacts Since 1925 . Chicago, IL: Council of State Governments.
_______________________________