تجزیه و تحلیل تعادل عمومی از روش اتصال بیمه برای تهدید آلودگی

This paper sheds additional light on the important issue of toxic accidental releases, or spills. The pollution control literature contains relatively little discussion of accidental releases. Even risk assessments are more about the risks to the public of predictable continuous emissions than about the risk of major accidents. Because spills of stored and transported chemicals are not continuous, or even inevitable, analogs of economic incentive approaches like emissions fees and transferable pollution allowances are inappropriate. Allowances are inappropriate because potential spills are often too harmful to allow, and a fee large enough to pay for clean-up and damages is beyond the means of many potential sources. The accidental release threat suggests an approach analogous to automobile liability insurance. Since a car can do more damage than most people can afford to pay for, states mandate insurance to ensure financial responsibility. Some potential sources of harmful spills are already required to demonstrate financial responsibility, and more could be. Insurance is one option for meeting financial responsibility requirements; in many cases, the only viable option. To develop testable hypotheses and determine where traditional partial equilibrium analyses would probably be misleading, the paper compares a very basic insurance bonding approach to the traditional command-and-control approach in a general equilibrium setting. That allows a comparison of effectiveness, and impacts on output levels, prices, and factor returns.

Spills, leaks, and improper disposal practices (hereafter, just ‘spills’) often make big headlines.2 Articles about routine emissions appear in the back pages. In contrast, economists have given the headline-grabbing accidental release of toxic chemicals comparatively little attention. And there is an extensive literature on the policy options for routine, more-or-less continuous emissions. Just as they dominate continuous emission policies, command-and-control (CaC) approaches aimed at minimizing scandals dominate the accidental releases policies. And there are many general reasons why there is usually a strong initial bias in favor of CaC approaches. Perhaps foremost among the sources of bias is that every citizen knows that the government can mandate and prohibit, but few have thought about incentive approaches. Most citizens expect a CaC approach. The authorities may be aware of policy options other than new constraints on behavior, but the incentives to adopt alternatives to the expected CaC approach are usually quite weak. CaC rules confer additional power that legislators and their administrators gladly accept. Noll (1996) described why much good advice “will appear to political actors as ridiculously impractical.” For example, legislators lean towards highly prescriptive legislation to avoid transferring power to the executive branch. In addition, a combination of economic illiteracy and perceived symbolism arouses some vocal opposition to market-based approaches. I have heard leaders of environmental organizations characterize emission fees as schemes that sell pollution rights with no environmental benefits because ‘big corporations’ will barely notice the cost, and thus not change their behavior. Some even equate marketable emission allowances with selling the right to commit crimes, including murder.3 The pro-CaC bias appears to be especially strong for catastrophic spills. Zeckhauser (1996) says that “large concentrated losses get substantially overweighted.” Such “risks create reputational externalities;” another name for politically damaging scandals. Biases persist, and inertia is a major factor, but market-based alternatives are getting harder to dismiss. The major shortcomings of CaC approaches are becoming more widely known. However, because of inertia, reforms were typically incremental policy additions rather than policy overhauls. For example, market-based policies were slowly grafted onto the CaC core of the air quality policy (Merrifield, 1990). Since bias is becoming a smaller problem than inertia, the outlook for market-based approaches is better as an initial response than as a CaC replacement. For example, a major overhaul of the Clean Air Act is not on the policy horizon, but market-based approaches are under serious consideration for the initial policy response to global warming concerns, and fear of stratospheric ozone depletion. This article aims to accelerate the entry of market-based policy options into the spills’ policy debate, and to improve the quality of that debate. It does so in two parts. The first three sections provide an informal overview of the many political and institutional obstacles to greater use of Insurance Bonding Approaches (IBA). Section 2 is a theoretical comparison of CaC and IBA. Section 3 is a literature review and it describes the policy setting. The remainder of the article provides a formal overview of general differences in the economic impacts of the IBA and CaC approaches. Because no model can address all of the issues raised in the first three sections, the second half of the article neglects some of the issues raised in first half. Section 4 outlines the general equilibrium model. A comparative static analysis and sensitivity analyses precede the summary and concluding remarks. A Appendix A contains most of the computations and derivations.

The results must be seen from the limited perspective that generated them. They are not as general as unambiguous comparative static results, but nearly so in light of the stability of the sensitivity analysis results. The model compared a technically efficient CaC approach to an efficient IBA. In the real world, for reasons discussed in Section 2, a technically efficient CaC policy is unlikely to ever be achieved. In addition, a comparative static analysis cannot capture dynamic benefits like the IBA advantages of private contract enforcement, and the increased freedom to innovate. Therefore, the analysis understates the likely advantages of an IBA. In the model, insurance coverage does not consume any labor or capital. The implied assumption is that insurance is a small enough sector of the economy to safely omit from the full employment equations. CaC administrative and enforcement costs are not included either. Since such CaC and IBA costs are likely to differ, that is another area for future model improvements. Despite its limitations, the model yields some significant insights, including an examination of major issues in a general equilibrium setting, and the identification of the key determinants of the results. There are significant differences in the impacts of the two approaches. The model indicates that even if the damage elasticities (the βs) are nearly zero, the IBA will still yield significant results at a small short-term cost to the economy. The exogenous variable (liability exposure) for the IBA also needs some more attention. Because many existing laws already establish some liability exposure, a change to 'full liability' is much less than a 100% change for many environmental threats. A legal analysis can establish the liability starting points that IBA could build upon. In addition, there may be a better way to express the IBA than a mandated increase in liability exposure. Since the effect of IBA depends most heavily upon the exposure elasticity of demand (μE), efforts to estimate μE are most likely to be worthwhile. The only major non-intuitive determinant of comparative static results was the wage elasticity of labor supply (σLL). To a much lesser extent, the price elasticity of demand for goods (α) was also important. That means that the partial equilibrium analyses that will be needed to explore the issues that a general equilibrium approach cannot address can produce reliable predictions if they at least include some labor market effects. Though the impediments to a nearly efficient IBA appear much less formidable than the CaC shortcomings that seem inherent in politics and governing, the IBA barriers and challenges are still formidable. Until they are addressed adequately, the default CaC policy will reign. The transfer of liability from a potential pollution source to an insurer raises several issues that have not yet been adequately addressed, either by industry innovation or through policy reform. The government must develop or acknowledge a method of calculating the mandated level of coverage, a way to revise calculations periodically, and determine each insurer's ability to pay potentially enormous claims. That means government officials must assess loss prevention efforts and monitor the assets of insurers and re-insurers. Such whistleblower roles seem less vulnerable to the inherent difficulties of governing than a detailed rulemaking role, but the public choice arguments are not yet widely understood. There should also be procedures for phasing out CaC regimes as formerly uninsurable risks become insurable.