Costs of air pollutants from shipping: a meta-regression analysis

Abstract

ABSTRACTThis study estimated the external cost of air pollution from shipping by means of a meta-regression analysis, which has not been made before. Three pollutants, which were included in most of the primary studies, were considered: nitrogen oxides (NOx), sulphur dioxides (SO2) and particulate matters with a diameter of max 2.5 micrometres (PM2.5). All primary studies included damages of health and a majority added impacts on agriculture and estimated the cost of air pollutants by transferring cost estimates from studies on costs of air emissions from transports in Europe. Different regression models and estimators were used and robust results were found of statistically significant emission elasticities of below one, i.e. total external costs increase by less than 1% when emissions increase by 1%. There was a small variation between the pollutants, with the highest elasticity for PM2.5 and lowest for NOx. Calculations of the marginal external cost of the pollutants showed the same pattern, with this cost being approximately six times higher for PM2.5 than for the other pollutants. Common to all pollutants was that the marginal external cost decreases when emission increases. Another robust result was a significant increase in the cost of studies published in journals compared with other publication outlets. These findings point out some caution when transferring constant external unit cost of air pollutant from shipping, which is much applied in the literature, and the cost functions estimated in this study could thus provide a complementary transfer mechanism.