4.1. Cost Savings from New Incentive Systems for Air Pollution Control
powered by 
This section uses two different methods for making a lower bound estimate of the potential additional savings that would be possible through a wider use of economic incentive mechanisms for the control of air emissions. The first approach is to review studies that model incentive applications. Of the some 25 studies summarized earlier in this report, the average ratio of CAC costs to incentive-based costs is 5.2, indicating that on average an incentive approach would cost about 20 percent as much as a command and control approach. It would be far too optimistic to assume that savings of this magnitude could be achieved unless improvements are incorporated into new and existing economic instruments. Indeed retrospective studies show that the actual savings from incentive applications have fallen well short of the theoretically predicted magnitudes due to higher transactions costs and greater regulatory hurdles than were anticipated. Consequently, one might more realistically assume that incentive-based mechanisms could save 50 percent over command and control approaches for stationary source air pollution control. Mobile source programs might realistically achieve one-half that percentage saving, primarily from greater reliance on incentives to reduce pollution from existing vehicles rather than new vehicles.
Table 3-2A of EPA (1990) estimates that air regulations will result in $42.8 billion expenditures by non-regulatory groups by the year 2000, $28.7 for stationary sources and $14.1 for mobile sources. Using the assumption of average savings of 50 percent in compliance costs for stationary sources and 25 percent for mobile sources, widespread use of incentive-based programs might reduce compliance costs by $14.4 billion for stationary sources and $3.5 billion for mobile sources in the year 2000. Some of these savings could be achieved through aggressive use of incentive provisions in the 1990 Clear Air Act Amendments. More realistically, new amendments that actively promote incentives rather than simply allow their use would be required to achieve these levels of savings.
An alternative method that can be used to check whether these estimates are reasonable is to identify some major areas where incentive-based mechanisms could be applied and calculate the potential savings for these applications. Of necessity, such an approach will neglect a host of relative minor applications whose cumulative impact on costs could nonetheless be quite large. Also, the discussion should not be taken as advocacy of any particular application, rather merely a rough indication of the magnitude of savings in compliance costs that may be achieved. Savings from incentives are estimated to be in the range of about $1.7 billion (possibly somewhat higher due to failure to calculate a few items), rising to a range of $4.1 billion by the year 2000. The following possible new applications could increase the annual savings by the year 2000 by an additional $10 billion:
- Allow trading between mobile and stationary sources. Evidence accumulated by the Congressional Budget Office (1982), Freeman (1982), and others suggests that, measured in terms of cost-effectiveness, mobile sources have been controlled more stringently than stationary sources. White (1982) estimated that a reallocation of control costs could have saved the equivalent of $240 per vehicle, or about $2.5 billion annually.
- The 1977 Clean Air Act Amendments effectively require all new power plants to scrub their stack emissions of sulfur dioxide. The Congressional Budget Office (CBO) estimates that if utilities were allowed to meet New Source Performance Standards (NSPS) requirements through the least cost approach (such as trading with other sources and use of low sulfur coal) the annual savings would be $4.2 billion per year.
- The Northeast states, Chicago, Houston, and other areas have considered creating trading regimes similar to RECLAIM. The USEPA recently proposed a NOx trading program for the Northeast states. The SCAQMD projected that RECLAIM would produce savings in compliance costs for stationary sources on the order of 65 percent in the early years of the program. Assume a 50 percent saving and apply that to EPA's estimate in Environmental Investments of the cost of controlling stationary source emissions of VOC and NOx. Controls for these pollutants comprise about 35 percent of all air pollution control outlays for stationary sources, or a projected $3 billion in 1992 and $4.5 billion in the year 2000. Savings of 50 percent of this would indicate potential savings in the range of $1.5 billion in 1992, rising to $2.2 billion in the year 2000 including $300 million estimated for RECLAIM in that year. EPA projected that about $500 million of these savings will be achieved with the recently proposed regional NOx trading program.
- Inspection and maintenance programs are less effective than they otherwise could be due to errors in detecting and repairing high emitting vehicles. Some of these "errors" are intentional, which is to say compliance with I&M requirements is poor in several programs. While EPA's new enhanced I/M requirements make improvements in the detection process and in dealing with evaporative system leaks, an incentive system based on taxing a vehicle's actual in-use emissions would have a number of advantages. For example, high-emitting vehicles might be scrapped if the owner had to pay for emissions. Under current and proposed EPA policy, the owner of such a vehicle could receive a waiver if an expensive repair would be required. The Oregon legislature has considered such an emissions tax. The potential savings might be estimated as follows: assume that vehicle hydrocarbon emissions, presently some 6.5 million tons annually, [OTA (1989).] could be reduced by 20 percent at a cost of $2,000 per ton, rather than the $5,000 per ton of a direct regulatory approach (the cost of basic I/M programs, estimated by EPA). The savings associated with an emissions tax could approximate $2 billion annually.