Government subsidy strategy selection in competing supply chains under green manufacturers' risk aversion

Abstract

This paper investigates how the government selects demand-side promotional subsidy strategies in a market where a green supply chain competes with a conventional supply chain and the green manufacturer is risk-averse. It focuses on analyzing the government's choice among no-subsidy, price-subsidy and the green-subsidy policies under different expenditure-benefit objectives. This study develops a game-theoretic model of competing supply chains, in which inter-chain competition is modeled as Cournot competition and intra-chain interaction as a manufacturer-led Stackelberg game. The green manufacturer's risk-averse behavior is characterized using a mean-variance model. A comparative policy analysis is conducted under equal-expenditure conditions and is further extended to a policy space with unequal expenditures. Under equal subsidy expenditure, the green-subsidy policy yields the highest environmental benefit, whereas the no-subsidy policy performs best in terms of economic benefit. When the government's comprehensive benefit is considered, the green-subsidy policy should be preferred at low subsidy levels. In contrast, the no-subsidy policy becomes more desirable at moderate or high subsidy levels. Under unequal subsidy expenditure, the optimal policy choice depends on the government's expenditure-benefit objectives and is determined by a policy space partitioned by iso-expenditure curves and iso-benefit curves. This study integrates competition between a green supply chain and a conventional supply chain, the green manufacturer's risk-averse behavior and the selection of demand-side promotional subsidy strategies into a single theoretical framework. By distinguishing between equal and unequal subsidy expenditures, this study derives the policy selection rules under different objectives and provides a clear basis for designing government subsidy schemes under fiscal constraints and market uncertainty.