TL;DR
This paper investigates how changes in expressway tolls affect traffic flow in Shanghai, China, using Electronic Toll Collection (ETC) data from 2019-2020. By employing Empirical Mode Decomposition (EMD) to filter out noise and isolate underlying trends, the study reveals that drivers of cars with seven seats or less become more sensitive to price changes as the travel distance and associated toll increase.
What Problem Does This Paper Address?
Accurate road pricing strategies are crucial for managing traffic congestion, reducing emissions, and improving urban mobility. A key component of designing effective road pricing is understanding the price elasticity of demand – how much traffic flow changes in response to toll adjustments. While previous research has explored this in various contexts, there's a need for localized studies that account for specific economic and geographical factors. This paper addresses this by focusing on Shanghai's expressways, aiming to quantify the price elasticity of demand for passenger cars (seven seats or less) and understand how it varies with toll amounts and travel distances. The study also aims to overcome challenges posed by data noise, such as the surge in ETC users and the impact of the COVID-19 pandemic.
Methods
The core of this research lies in its data processing and analysis techniques. The authors utilize Empirical Mode Decomposition (EMD), a method well-suited for analyzing non-linear and non-stationary time series data. EMD decomposes a signal into a set of intrinsic mode functions (IMFs) and a residual, effectively separating different frequency components and trends. This is crucial for isolating the impact of toll changes from other confounding factors like the growth in ETC users and pandemic-related disruptions.
After EMD processing, the study calculates demand price elasticity using the midpoint formula of arc elasticity. This formula is chosen to provide more robust results, especially when dealing with price changes. The analysis then proceeds by:
- Data Normalization: Adjusting data to account for factors like the growth of ETC users.
- OD Pair Analysis: Grouping traffic data by origin-destination (OD) pairs to analyze specific routes and their corresponding charges and traffic flows.
- Filtering Outliers: Excluding OD pairs with minimal charge differences or positive elasticity values, which might indicate the influence of unaddressed factors.
- Regression Analysis: Applying linear, logarithmic, quadratic, and exponential models to explore the relationship between charge (distance) and the absolute value of demand price elasticity.
Datasets
The study uses measured Electronic Toll Collection (ETC) data from Shanghai expressways for the period of 2019 to 2020. This dataset includes information on vehicle entry/exit stations, charge amounts, and vehicle types. The focus is specifically on cars with seven seats or less. The data was collected to capture traffic flow before and after a change in the expressway charging standard implemented on January 1, 2020.
Key Insights and Ideas
- EMD as a Noise Filter: The application of EMD is a key methodological contribution, effectively separating the underlying trend of traffic flow from noise caused by factors like ETC user growth and external events (e.g., COVID-19). This allows for a more accurate assessment of price elasticity.
- Distance-Dependent Elasticity: The study strongly suggests that demand price elasticity is not uniform across all travel distances. As the travel distance (and consequently, the toll) increases, drivers of passenger cars become more sensitive to price changes.
- Targeted Road Pricing: The findings imply that road pricing strategies might be more effective for longer-distance, higher-toll routes. This offers valuable insights for policymakers aiming to optimize congestion management and revenue generation.
- Sensitivity to Small Price Changes: The research highlights that for very small charge differences (less than 1 yuan), the calculated price elasticity might be less accurate, suggesting that significant price adjustments are needed for a clear observable impact on demand.
Key Results
- The study found that the demand elasticity of expressway charges is related to the number of charges (distance). Specifically, longer travel distances lead to higher charges and a greater absolute value of demand price elasticity.
- Road users are observed to be more sensitive to price changes when facing higher road charges.
- The quadratic model provided the best fit for the relationship between charge and elasticity value, indicating a non-linear relationship.
- The study identified that for charges greater than 10.31 yuan, the elasticity value is greater than 1, signifying elastic demand.
Limitations and Future Work
- Exclusion of Positive Elasticity Pairs: The paper acknowledges that two OD pairs with positive demand price elasticity were excluded, suggesting that other influencing factors might have been at play for those specific routes. The authors note that these exclusions were made to prevent these factors from affecting the research results.
- Data Granularity: While the study focuses on cars with seven seats or less, the change in vehicle classification standards made it difficult to accurately distinguish all vehicle types.
- Small Charge Differences: The accuracy of elasticity calculations may be compromised when charge differences are very small (e.g., less than 1 yuan).
- Future Work: The paper suggests that its findings can provide references for other research on demand price elasticity. Further investigation into the specific "other influencing factors" that lead to positive elasticity in certain cases could be a valuable direction. Additionally, exploring the elasticity for different vehicle types beyond passenger cars would offer a more comprehensive understanding.
Summary updated