Skip Navigation Planning & Markets
Subscribe Submission Requirements Editorial Board Archive Links Search Home



On February 23, 1993 (Federal Register), the US Environmental Protection Agency (EPA) issued guidelines for states that want to award mobile emission reduction credits (MERC) when clunkers are retired. The EPA's guidelines focus on the legal issues that govern the generation and use of MERCs. The EPA did not suggest specific EMVR policies, but they flagged potential program design issues. The state policies suggest procedural details. For the sake of brevity, the discussion is limited to the most noteworthy policy features. Three states (California, Illinois, and Texas) have EMVR policies or policy proposals, and have conducted EMVR experiments.

The federal MERC guidelines (Federal Register, February 12, 1993) further illustrate how the Command-and-Control core of the CAAA limits the potential environmental and economic benefits of emission trading opportunities. The federal and state MERC policies are new examples of the drawbacks of an evolutionary approach (Merrifield 1990) to environmental policy (creating new programs as add-ons to existing ones).

Stationary sources cannot use MERCs to avoid or delay compliance with the CAAA's strictest technology-based rules (Best Available Control Technology and Lowest Achievable Emissions Rate [LAER], p 111-35). That severely limits the use of MERCs to improve cost-effectiveness and shrink marginal cost gaps. Such restrictions are inefficient, and in non-attainment areas, where new sources must more than offset their emissions, the LAER requirement is environmentally counter-productive.

MERCs can be used to delay compliance with RACT (Reasonably Available Control Technology) or NSR (New Source Review) rules. It's unclear whether a steady stream of MERCs could be used to permanently avoid RACT or NSR compliance. The rules tentatively allow MERCs not used in one time period to be used in another. Allowing MERCs to be banked for later, or borrowed from the future, makes sense if marginal damages are constant, or if the marginal benefits of leveling ups and downs in MERC availability outweigh the effect of any negative differences in marginal damages. Inter-temporal swapping could beneficially shift emissions from daily peaks to off-peak times (more on this later).

California and Illinois envision only sporadic EMVR efforts. MERC banking in Texas allows for continuous EMVR by dealers expecting a continuous demand for MERCs. Illinois suggested a $3/car oversight charge. In California, SCAQMD’s (South Coast Air Quality Management District) Director must approve each EMVR episode.

In Illinois and Texas, a retirement creates MERCs that last 2-3 years. In California, MERCs, which SCAQMD calls MSERCs, are good for five years, though cars must only ‘demonstrate’ a probable minimum remaining life of three years. Nitrous oxide (NOX) and volatile organic compound (VOC: an ozone precursor) credits are marketable. Illinois’ policy recommendation does not include MERC uses. Texas suggests the private use of MERCs for compliance problems like brief production increases, delayed installation of pollution control equipment, and for the CAAA requirements that apply to non-emitters. SCAQMD in California, prescribes several narrowly defined uses (specific industrial processes) for MERCs, and businesses, including non-emitters subject to employee trip reduction requirements (Wachs and Giuliano 1992), can help fund SCAQMD’s air quality investment program in place of compliance with some rules. EMVRs are an explicitly approved use of those funds. SCAQMD may receive additional funds for EMVR efforts through the state Air Resources Board’s proposed Measure M-1. The goal of M-1 is to retire up to 75,000 older vehicles between 1999-2010, and lesser amounts in the next couple of years to gain experience.

Local governments may buy MERCs to help show progress towards goals specified in state implementation plans. Local governments in areas nearing non-attainment status may see EMVR as a good way to help avoid a non-attainment designation.

The fleet average is the figure used to compute the size of MERCs, even though not every retired car is replaced (3). In addition, unless the EMVRs inflate used car prices enough to cause sugnificant importing from outside the target airsheds, the average replacement car at the end of the replacement domino effect is newer than the fleet average. The CAAA's environmental bonus requirements and attrition adjustments (4) (as high as California’s 20%/year) build in additional conservatism. Because of selection bias (5), MERCs are given a shorter life than a clunker’s average expected lifetime of six years. In Illinois, a clunker’s estimated MERC value includes a tampering factor.

In Illinois and Texas, EMVR was grafted onto an existing inspection and maintenance (I&M) program. An EMVR produces a MERC only after two emission tests. A test #1 failure forces a car owner to make repairs (at least enough to qualify for a waiver ) (6) or retire the car. The second test determines the car's MERC value. In Texas, the MERC value also depends on a repair estimate. SCAQMD grants MERCs that vary with vintage on the basis of an emissions model. Though California SMOG CHECK rules probably help clunker owners choose retirement, the only motive cited in the SCAQMD rules is the cash offer of the EMVR sponsor.

In Texas and California, cars must be insured, operable, and registered in the target area to be eligible. Illinois only seems to require a title check. IEPA (Illinois Environmental Protection Agency) said the spring is the best time for an EMVR. Cars used in the winter probably have a lot of remaining life. A mechanical screening is part of Illinois' recommended rules, but a car's condition does not influence the expiration date of the MERCs it represents. California’s Bureau of Automotive Repair (BAR, 1994) conducted a pilot project in the Sacramento area that varied the price paid with measured emissions and an estimate of the car's remaining life.

Since newer cars have much more efficient emission controls, Illinois’ and SCAQMD’s rules require that cars be 1981 models or older. Unless the newer cars' emission control systems decay more rapidly than other parts, there is no reason to ever update the model year constraint. EMVRs and attrition will gradually make EMVRs obsolete. The difference between the emissions of post-1981 clunkers not worth repairing and average replacement cars won’t be large enough to generate significant MERCs. Consider, for example, TNRCC (Texas Natural Resource Conservation Commission) data for Harris County (nearly all of Houston), Texas. The number of 1981 and older cars fell from 573,962 in 1989 to 282,921 in 1993; a 19.3% annual rate of decline. Further attrition at that rate, plus about a dozen UNOCAL-sized (1991) EMVR episodes would virtually eliminate the supply of eligible clunkers by the year 2000.

In most cases, vehicle characteristics did not affect price because of concern about tampering. The only exceptions were the Illinois Pilot Project, where prices varied by model year, and the California’s Bureau of Automotive Repair Pilot Project (1994), where the EMVR cash offers surprised participating motorists. Surprise, a key factor, is lost once EMVR goes beyond a pilot project. An early Texas proposal, later rejected as possibly too speculative, inequitable, and complex, linked the price to repair costs and insurance company data for replacement values.

Paying the same price for each vehicle substitutes selection bias for the tampering incentive (Alberini et al 1994, 1995). EMVR programs' tendency to attract the least driven vehicles with the fewest remaining miles is especially severe when the price paid is low.

Such pricing policies may cause other problems. The context of EMVR (evolving interpretations of the 1990 CAAA) means that MERC prices only reflect their usefulness in easing the burdens the CAAA imposes on local governments and businesses (including non-emitters). Similarly, Anderson and Howitt (1995) found that State Implementation Planners counted only certain air quality benefits of transportation policies. MERC prices will not reflect the many other benefits to society that accrue from an EMVR (Plotkin 1992). Older cars are not as fuel efficient, or as safe, as the newer cars likely to replace them. MERC prices also won’t reflect the social benefits of lower levels of pollutants already below legal limits, or the probable benefits of evening out emissions by time of day. Swapping mobile for stationary emissions (except perhaps for electric utilities) should shift some emissions away from the morning and evening peaks that produce the worst ambient air quality, to other times of day. Society also benefits from replacement lags, and when some non-replacement reduces traffic. Those external benefits justify government subsidization of accelerated vehicle retirement.

Another so far unnoticed potential benefit of an EMVR program (especially a continuous one) is that it could make some regional pollution offset markets more competitive. Without MERC sources, pollution sources are few enough in some areas to cause monopoly behavior, including higher prices and the hoarding of credits to limit competition. In some places, the MERCs could make emission offset banking and inter-firm offset trading viable.

The spatial redistribution of emissions may or may not be desirable. It depends on the location of the ambient air quality monitors, and the spatial distribution of people. The latter is also a fairness issue. Low-income housing is often concentrated near stationary sources that may use EMVRs to avoid emission cuts. Large-scale EMVRs could also disproportionately impact low-income households by raising the price of used vehicles.

The tone of state and federal guidelines and recommendations is very tentative. Their authors recognize that additional data and thoughtful analysis of existing evidence can produce some significant insights. For example, Illinois’ authorities say that an emissions model (7) will eventually replace emissions testing as the basis of MERC calculations.

page 2



USC Seal

Main Page | Subscribe | Submission Requirements | Editorial Board | Archive | Links

ISSN 1548-6036

Copyright 1999-2000
University of Southern California
Los Angeles, California 90089-0626