Abstract: In recent years, the carbon emission coupled with production systems has been gaining more attention from both industrial practitioners and academic scholars. To curb carbon emission, cap-and-trade scheme is often utilized by government as a cost effective mechanism to reduce carbon emission without sacrificing economic robustness. In order to calculate free carbon permits, two types of historical-based methods are often adopted, i.e. grand-fathering and benchmarking. The grand-fathering approach is used to allocate the free carbon allowance (cap) according to historical total emission volume on a yearly basis. In contrast, the benchmarking approach cap is calculated on the basis of industrial best-practice carbon emission per production unit. In this paper, a two-echelon supply chain consisting of a dominant supplier and a manufacturer where they run a Stackelburg game is investigated. Then the impact of the two approaches upon production decisions such as optimal order and production quantity, per unit volume of carbon emission abatement (PCEA) and wholesale price is analyzed. Taking carbon conscious factor of market demand into consideration, news vendor models in four scenarios in which the cap-and-trade policy affects the upstream or downstream players respectively are solved. The results show that the benchmarking approach, although is intuitively viewed as a more rigorous method in terms of environmental constraint, encourages the downstream player to improve order quantity, i.e. the optimal order quantity is bigger by the constraint of benchmarking method. Meanwhile, the PCEA is higher under benchmarking approach than it is under grand-fathering method, therefore the total volume of carbon emission abatement is higher under benchmarking scheme. In addition, it is suggested that the downstream player will increase order quantity under benchmarking scheme regardless of the supply chain's node being affected by the cap-and-trade policy. When the market demand is carbon conscious, i.e. linearly related to PCEA, the downstream player's order quantity will depend on PCEA even without the constraint of cap-and-trade. Finally, managerial insights are obtained, which include (1) the government use benchmarking approach to allocate cap is a better choice in terms of improving the total carbon emission abatement volume; (2) the benchmarking approach is not always effective under some circumstances where it is affected by carbon trade price, carbon technical coefficient and stochastic factors etc.

Key words: low-carbon supply chain, news-vendor model, market demand of carbon sensitivity, carbon control policy