تعیین اندازه مقرون به صرفه یک منطقه انتشار تجاری برای دستیابی به استاندارد محیط

Little effort has been devoted to determining the cost-effective trading region for emission permits that achieve an ambient standard. With regional expansion, control by more distant, lower marginal cost sources supplants control by higher marginal cost sources in the original region. However, the impact of more distant source emissions on ambient receptors will typically decline. The aggregate effect on costs of this tradeoff depends crucially on the standard’s stringency. Since a rule-of-thumb for the effect of regional expansion requires overly-restrictive assumptions, we model nitrogen oxide permits subject to nitrogen loading standards for Chesapeake Bay. Regional expansion substantially reduces cost only when the standard is stringent.

Even though the US has carried out the most extensive and diverse experimentation with systems of tradable emission permits of any country, the systems in use share a fundamental characteristic: they make minimal or no use of the differential ambient impacts of sources to optimize system design. Thus, permits must be purchased for source emissions rather than their degradation of air quality as measured by ambient monitors. Emission-based systems are used to improve air quality even when the location of an emissions source is known to cause significant and non-uniform ambient degradation, in particular, tropospheric ozone and acid deposition. Momentum will likely produce efforts to employ emission-based systems to address important remaining regional air pollution problems in the US: fine particulate matter, regional haze (which also is caused by fine particulate matter), and the 8 h standard for tropospheric ozone. In a few cases, emission-based systems partially incorporate the differential ambient impacts of emissions by restricting trades to specific geographical regions. Such restrictions are present in, for example, the Federal non-attainment new source review program and the RECLAIM program in southern California. The Clean Air Act ((AA) requires compensating emission reductions (offsets) for increased emissions from new or modified “major” stationary sources in non-attainment areas (42 USC, Section 7503). Offsets may be obtained from other sources in the same non-attainment area, and, in general, states may allow offsets to be obtained from sources in another non-attainment area if “the other area has an equal or higher non-attainment classification than the area in which the source is located” and “emissions from such other area contribute to a violation of the national ambient air quality standard in the non-attainment area in which the source is located” (42 USC, Section 7503(c)). For emissions of volatile organic compounds and oxides of nitrogen (NOx), the related “Emission Offset Interpretive Ruling” conveys the Environmental Protection Agency (EPA)’s recommendation that the compensating emission reduction should exceed the emission increase from the new or modified source (40 CFR 51, Appendix S IV(D)).1 This recommendation goes even further by stating that except when provided by adjacent facilities, offsets of sulfur dioxide (SO2), particulate matter, and carbon monoxide should be justified on the basis of “atmospheric simulation modeling to ensure that the emission offsets provide a positive net air quality benefit” (40 CFR 51, Appendix S IV(D)). The RECLAIM program, which comprises two separate emission trading programs for NOx and oxides of sulfur (SOx), divides the South Coast Air Quality Management District into two trading zones for the purpose of new source review. The cities on and nearest the coast are in Trading Zone 1. Inland, downwind cities are in Trading Zone 2. All emissions from a new or relocated facility in Trading Zone 1 must be offset by emission reductions (RECLAIM Trading Credits) occurring at facilities that also are in Trading Zone 1; a new or relocated facility in Trading Zone 2 may obtain offsets from facilities in either zone (Rule 2005, Section (e)). The geographical restriction on trades that is present in the RECLAIM program is atypical of the “second generation” emission trading systems, i.e., the regional and national cap-and-trade systems that were created after the Federal new source review program. For example, neither the Federal “Acid Rain” SO2 emission trading program (created by the 1990 amendments to the Clean Air Act) nor the Ozone Transport Commission’s summertime NOx emission trading program (created in the mid-1990s) contain any geographical restrictions on trades. Many aspects of emissions trading regions or zones have been examined empirically with simulation models, where trading is allowed within but not between zones. Atkinson (1983) considers the impact on the cost-effectiveness of a local SO2 permit trading system, when it is enlarged to include receptors of long-range acid deposition. Also, Atkinson and Tietenberg (1982) examine the cost-effectiveness of zonal trading systems where the control authority is not able to make an initial allocation of control responsibility that would assure regional, and hence inter-zonal, cost-minimization to achieve ambient standards. Since trades are prohibited across zonal boundaries, the initial allocation affects the total cost of control to meet ambient standards. For the region studied by Atkinson and Tietenberg (1982), the increase in total control costs is substantial relative to a system without zones. However, as Tietenberg (1995) points out, overly large zones can cause over-control of distant sources, so that control costs are unnecessarily high to achieve an ambient standard. A study by ICF Resources Inc. (1989) investigates the costs of incrementally increasing the size of a trading region. They expand its size from the plant level—so that only discharge points within a plant can trade emission permits—to plants within a state, and finally to plants within a region. Total control costs to achieve ambient SOx standards decline (at a decreasing rate) as the region is expanded. All of these studies investigate control programs that address the effect of atmospheric emissions on the attainment of ambient air quality standards or on long-range acid deposition. In contrast, this paper investigates the optimal size of the trading region for atmospheric emissions as they affect aquatic nitrogen loading. More importantly, none of the previous studies explicitly analyzes the tradeoff encountered in the regional expansion of an emissions permit system: control will be shifted to lower marginal cost sources, but their emission reductions may yield a smaller ambient improvement. With regional expansion, Far sources (those in the annexed region) will undertake control in place of some of the Near sources (those in the original region), if the marginal costs of the former are smaller than those of the latter. Yet this expansion may have costly unintended consequences. Since the impact of emissions on centrally-located receptors, measured using source–receptor transfer coefficients, is typically smaller for distant than for local sources, this substitution may increase total control costs. This tradeoff depends critically on the ambient standard, which determines the marginal cost of control. Is there a rule-of-thumb which can help us easily determine which of these effects will dominate? We show that such a rule exists only for a highly stylized case with a constant and common marginal control cost. The rule-of-thumb states that, given constant and common marginal control costs in each of two regions, total control costs for an emissions permit system used to achieve an ambient standard will decrease if the ratio of the original emission permit price to the price after regional expansion is greater than the ratio of the sum of the transfer coefficients for the Near sources to that for the Far sources.2 While the restrictive assumptions required for this rule-of-thumb are unlikely to hold in reality, relaxing them renders our rule-of-thumb inoperable and results in a significantly more complex theorem. Thus, one must typically determine the expected effect of regional expansion via simulation modeling. We then utilize a linear programming model to determine the extent to which expansion of a hypothetical NOx emission trading region affects total costs for different levels of ambient improvement. Our initial region includes major utility and non-utility point sources in the Chesapeake Bay Area (defined further). The expanded region includes these sources and similar sources in the states surrounding the Chesapeake Bay Area, termed the Chesapeake Bay Airshed (defined further). Our model uses estimated marginal costs for NOx emission reduction and transfer coefficients linking emissions to delivery of nitrogen to the Bay, the latter resulting in reduced water quality and ecosystem integrity. We first model an emissions permit trading program which must meet nitrogen loading standards for Chesapeake Bay. As a baseline we also model an ambient permit system, where nitrogen loading permits are purchased in order to satisfy ambient standards. By increasing the geographical size of the emissions trading region, we determine whether control costs increase or decrease for the emissions permit system given different levels of allowable loading of the Bay. We compare these control costs with those for the ambient permit system, whose total costs of control decline as the region increases. Finally, we compute shadow prices for both permit systems.3 Our results indicate that the total costs of an emissions permit system for the Bay Area and Chesapeake Bay Airshed trading regions depend dramatically on the required level of ambient improvement. For a relatively unstringent ambient standard, regional expansion substantially increases the total costs of the emission permit system. However, for a considerably more stringent ambient standard, regional expansion substantially reduces total control costs.4 The remainder of the paper is organized as follows. In Section 2, we present the background for our investigation of NOx permit trading in the Chesapeake Bay Area. In Section 3, we develop an example of regional expansion which reduces total control costs for a moderate ambient standard, but increases these costs for a more stringent ambient standard. We then derive a rule-of-thumb theorem. Section 4 develops our simulation model. Our data and results follow in 5 and 6. We draw conclusions in Section 7.