<p>At present, no research has been able to answer the questions of how to design effective government subsidy and the optimal altruistic object mode to promote the long-term sustainable development of the low-carbon closed-loop supply chain (LC-CLSC). To answer the above questions and achieve the research objectives of this paper, this paper considers the altruistic behavior of leading enterprise under government subsidy, and uses differential game research method to construct dynamic decision-making models under multiple altruistic object modes from a long-term dynamic perspective, and obtains the feedback dynamic equilibrium strategies from all parties, and reveals the optimal design of the government subsidy and the optimal altruistic object mode. The research results show that: First, the government's optimal subsidy rate for emission reduction and recycling should be reduced as the manufacturer’s altruism towards the retailer or recycler intensifies, and this move is also conducive to an increase in consumer surplus and social welfare. Second, if the manufacturer shows the same altruistic degree to both the retailer and recycler, its optimal altruistic mode is that it should first choose to be altruistic to both parties, then to be altruistic only to the retailer, and finally to be altruistic only to the recycler, which is its worst choice. Finally, whether the manufacturer shows altruism towards the retailer or recycler, the key to the long-term sustainable development of LC-CLSC lies in its discount rate and the altruistic degree, as well as its mutual collaboration with government subsidy. This paper has significant implications that it can provide theoretical references for enterprises of the LC-CLSC to select reasonable altruistic object modes to carry out scientific altruistic practices, and offers suggestions for the government to formulate emission reduction and recycling subsidy policies.</p>