Modeling of GHG Mitigation Strategies in the Trucking Sector
- Author(s): Guerrero, Sebastian E.
- Advisor(s): Madanat, Samer M.
- Leachman, Robert C
- et al.
In response to the growing climate change problem, many governments around the world are seeking ways to reduce the greenhouse gas (GHG) emissions of various sectors of the economy. The trucking sector is important in meeting this challenge in the US because it is responsible for a share of emissions that is significant and rapidly growing. For governments to intervene in this sector smartly, they need models that capture its key incentives, constraints and dynamics, while making the most out of the limited data available. However, existing models fall short of this ideal. This dissertation first introduces the Trucking Sector Optimization Model (TSO) as a tool for studying the decisions that carriers and shippers make within a short-run time horizon--modeling the dynamics of truck fleets, penetration rate of Fuel Saving Technologies (FSTs) such as aerodynamic improvements and low rolling resistance tires, and changes in the demand of trucking. In addition to estimating tailpipe GHG emissions, the model also estimates emissions from upstream fuel production sources, vehicle manufacturing, and pavement rehabilitation activities.
This model is then used to evaluate the effectiveness of various incentives-based and regulation-based strategies that California's government could implement in the trucking sector to help achieve the objectives of the Global Warming Solutions Act of 2006 (AB 32). The strategies analyzed are: fuel taxation, mileage taxation, truck purchase taxation, FST subsidies, FST regulations, increases in the allowed weight of trucks, and the Low Carbon Fuel Standard recently introduced in California. Results indicate that there presently exist significant economic incentives for carriers to invest in FSTs beyond what is currently commonplace. The correction of market mechanisms that are responsible for this apparently suboptimal behavior, would lead to significant reductions in emissions, and would also allow for incentive-based strategies to have their first-best outcomes. Without making these corrections, the regulation approach currently adopted in California, of mandating certain investments in FSTs, serves as a reasonable first-step in meeting AB 32's medium-term emissions target. However, moving forward, the correction of these market mechanisms and subsequent implementation of incentives-based strategies, particularly those that are complementary with each other, should be a priority. Based on their estimated effectiveness, these and other recommendations are articulated in a seven-step plan for reducing trucking related emissions in the state.
The remaining chapters of this work study some long-run factors that affect how carriers manage their fleets and invest in FSTs, in particular considering that they often discount heavily the future because of the existence of various market failures, hidden costs and uncertainties in the industry. The nature of these issues is not investigated deeply in this research, but their effect on carriers is captured by parameterizing the level of discounting in an improved model called the Trucking Sector Trip Segmentation Model (TSTS). This model represents the long-term decisions made in this sector better than the TSO model by: (i) modeling endogenously how trucks are utilized throughout their service-lives, and (ii) capturing some heterogeneity in truck retirements. The first of these improvements is made possible by incorporating information on the performance of trucking (the ability of carriers to complete shipments) and on the spatial distribution of shipment demand. The second of these improvements is made possible by assuming that truck retirements follow a log-logistic function. Combining both of these methodological improvements with a parameterized discount rate provides analysts a more flexible model for studying the long-term decisions made in the trucking sector, especially regarding FST investments, which impact greatly emissions and costs.
The TSTS model is then used to evaluate the effectiveness of three additional governmental interventions that reduce GHG emissions, which could not have been studied with the TSO model. Improvements in trucking performance--by reducing congestion or shipment waiting times for example--were found to significantly incentivize investments in FSTs and reduce GHG emissions. However, 40 - 50% of these reductions were offset in the aggregate by increases in the demand for shipments precipitated by the lower market prices of trucking. Mode-shifts were also found to incentivize investments in FSTs because they distort the spatial distribution of shipments in ways that favor making greater capital investments because trucks are used more intensely and retired quicker. And finally, implementing FST regulations that only apply to a subset of the truck fleet (as in California currently) also reduces emissions, but incentivizes other changes in how the industry operates.
The TSO model is best suited for studying the dynamics and transitions of truck fleets in response to governmental interventions, while the TSTS model is best suited for studying long-run responses. Together, they allow policy makers and researchers to study a wide range of issues in the trucking sector, considering many interactions and responses that had not been adequately explored previously. They also share a rich theoretical framework that can be used in future research to develop better models of this sector, especially to help design interventions that have environmental objectives.