Dissertation/Thesis Abstract

From Detection Through Repair: Vehicle Emissions Testing in the San Joaquin Valley
by Wimberger, Emily R., Ph.D., University of California, Davis, 2011, 93; 3482289
Abstract (Summary)

Vehicle emissions testing, otherwise known as Inspection and Maintenance (I/M), is prevalent in areas of the United States with poor ambient air quality, because tailpipe emissions are the largest single source of ozone-forming pollutants. Even though California has the most stringent emissions requirements and the largest I/M program in the nation, as of 2011, nine of California's fifteen Air Basins failed to meet the ground-level ozone National Ambient Air Quality Standards.

Where the official California Smog Check program has failed to reduce tailpipe emissions sufficiently, private organizations have stepped in. One such organization, Valley Clean Air Now (Valley CAN), a non-profit advocacy group, has offered through its Tune In & Tune Up (TI&TU) Program free emissions-testing and up to $500, per vehicle, towards emissions-related repairs. Valley CAN targets vehicle owners in the San Joaquin Valley who may be driving while unregistered because of the expense of proper repairs.

In this dissertation, I use those vehicles that attended TI&TU events from 2005 through 2009 to analyze three aspects of vehicle emissions testing in California: the detection of gross polluting vehicles, the Valley CAN program's cost-effectiveness, and the durability of emissions-related repairs. For each of the 2,048 vehicles, I have cross-sectional data that includes detailed vehicle-specific characteristics, emissions test results, registration history, and the emissions-related repair cost.

Vehicles participating in the TI&TU program are an unusual subset within the California fleet. The largely homogenous TI&TU fleet tends to be older and to have higher emissions than the typical vehicle in California. The majority of TI&TU vehicles are also not in compliance with DMV registration requirements. Thus, the TI&TU vehicles allow for the analysis of unregistered and illegally driven vehicles, vehicles that are often ignored in the literature even though they pollute heavily.

Shortfalls with the California Smog Check program have been attributed, in part, to gross polluters, namely the small percentage of the vehicle fleet that emits at more than twice the regulatory standard. California tries to identify gross polluters, who are then directed to special Smog Check stations, with what is called the High Emitter Profile (HEP). As the full HEP specification is proprietary, I approximate that statistical model and measure its accuracy when applied to vehicles participating in the TI&TU event held in Bakersfield in 2009. My findings, reported in Chapter 2, suggest that although the HEP model is accurate overall, it is not well calibrated to identify gross polluters specifically. In Chapter 2, I also estimate a model in which gross polluters are identified only by model year and find this to be a close approximation to the HEP, with a much smaller informational burden and cost to California.

Only registered vehicles are subjected to the HEP, which is applied in conjunction with DMV renewal. At the time of the TI&TU event in Bakersfield, merely 21% of the 211 vehicles had the proper registration status and were being driven legally. Restricting the statistical comparison to these few registered vehicles, I find that neither the HEP nor the model-year approximation identifies any gross polluters.

In Chapter 3, I exploit the repair cost and emissions data to estimate a marginal abatement cost schedule for vehicles participating in two TI&TU events held in Bakersfield in 2006 and 2009. I find that 1995 and older model year vehicles have a lower abatement cost than newer vehicles across all emissions levels. Because older vehicles pollute more, an optimal allocation of repair funds should target those vehicles. Total emissions would be reduced by an estimated 22% were the program to shift from the actual flat $500 voucher to the first-best voucher scheme, which would be vehicle-specific. Even a two-tier voucher based on model year would yield an estimated 18% decrease in emissions. The estimated abatement cost schedule also indicates that optimally redistributing the expenditures required to bring each vehicle to California Smog Check standards but no further could reduce emissions by an estimated 27%.

In Chapter 4, I focus on emissions-related repairs, which are ultimately responsible for reducing tailpipe emissions. Most ex ante and ex post cost-benefit studies pertaining to I/M programs assume, implicitly or explicitly, that emissions-related repairs decrease vehicle tailpipe emissions without affecting the rate at which emissions evolve over vehicle usage. I test this assumption using data on emissions and repairs for a fleet of vehicles participating in the TI&TU program from 2005 through 2009.

I find that emissions-related repairs do affect the rate at which a vehicle emits tailpipe pollutants over usage (defined by the mileage increment between inspections). This result implies that the measure of abatement benefits assumed in cost-benefit studies may be inappropriate. I also find that the effect of repairs on the accumulation of emissions over usage depends on the model year. Repairs conducted on vehicles older than the 1993 model year tend not to be durable, while repairs conducted on 1994 and newer model years have a preventative effect. That is, for these newer vehicles, repairs reduce the emissions accumulated over usage compared to what emissions would be in the absence of the repairs. Although these findings call into question the results of Chapter 3, the discrepancy in repair durability is not enough to counteract the effectiveness of repairs targeting older vehicles.

Although vehicles attending the TI&TU events are not representative of the vehicles subjected to the California Smog Check program, this dissertation identifies areas in need of improvement within the current program structure. The next step is to apply the methodologies developed in this dissertation to the much larger population of vehicles subject to the California Smog Check program.

Indexing (document details)
Advisor: Williams, Jeffrey C.
Commitee: Larson, Douglas M., Merel, Pierre R.
School: University of California, Davis
Department: Agricultural and Resource Economics
School Location: United States -- California
Source: DAI-A 73/03, Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Environmental economics, Economics
Keywords: California, Emissions testing, Latent class model, San Joaquin Valley, Vehicle emissions
Publication Number: 3482289
ISBN: 9781267029454