Association for Public Policy Analysis & Management

Maritime transportation plays an essential role in international trade, but is also a major source of pollution, largely due to low fuel quality and poor emission control. Emissions from oceangoing ships have been estimated to cause 60,000 deaths globally and 14,500-37,500 deaths in East Asia (Corbett et al., 2007; Liu et al., 2016) annually. A popular measure to control ship emissions and associated impacts is fuel content standards, enforced in Europe, North America, and China by setting up Emission Control Areas (ECAs). Whether and to what extent these ECAs can mitigate environmental and health impacts from ship emissions, however, are not yet investigated by ex post evaluations.

We provide an evaluation of the air quality and health effects of ECA fuel standards at the main ports of China’s Yangtze River Delta (YRD). There were no fuel content regulations in China, except the 3.5% sulfur content cap set up by the International Maritime Organization globally. Following the national policy agenda to set up ECAs, the YRD region was the first to implement a cap of 0.5% sulfur content for ships at its main ports on April 1, 2016. The area includes the Port of Shanghai, the world’s largest port by container volume and second largest by cargo tonnage, and the Port of Ningbo-Zhoushan, the world’s largest by cargo tonnage and fourth by container volume. A phase-in period of two months was introduced at the Ningbo-Zhoushan port for more flexibility.

Our research connects to the literature that evaluates policy effectiveness on addressing transportation pollution and strategic compliance in response to regulatory flexibility on the one hand (Auffhammer & Kellogg, 2011; Davis, 2008). On the other hand, it infers the spatial extension of negative externalities from oceangoing ships, by considering ECA implementation as a natural experiment. Different from the focus on local externalities from transportation in the literature (e.g., Knittel, Miller, & Sanders, 2016; Moretti & Neidell, 2011; Schlenker & Walker, 2016), we are able to investigate impacts of a longer distance.

We compile the air pollution data from monitoring stations and weather conditions from weather stations and match them by distance. Besides the traditional identification relying on the temporal difference for a regression discontinuity design (e.g. Davis, 2008), we also use wind patterns to determine whether monitoring stations are affected by pollution created at the ports. We follow a previous method to construct the influences from wind direction and speed (Barwick, Li, Rao, & Zahur, 2018). This helps us identify the effect of ports more reliably.

Results show that the more stringent fuel content regulation led to 20-30% reduction in SO₂ concentration, 10-30% reduction in PM_2.5, and 10-25% reduction in PM₁₀ in 100-kilometer radius of the two major ports. The effect remains significant in areas 100-200 kilometers away from the ports. The phase-in period at the Ningbo-Zhoushan port did not cause the policy benefits any smaller than that of the Shanghai port. A back-of-the-envelope calculation (World Health Organization, 2016) suggests that the avoided premature death would be more than 2,000 persons annually.