This paper investigates the optimal carbon allowance procurement strategy for a regulated firm operating under an Emissions Trading Scheme (ETS) to achieve compliance in a net-zero world. We model a regulated firm that must hedge the carbon emission exposure arising from fulfilling uncertain market demand in a make-to-order environment. Within this combined procurement framework, the firm determines an optimal quantity of allowances to secure via forward contracts before this demand uncertainty is resolved, while retaining the flexibility to cover any deficit in the spot market, which is subject to its own liquidity and price risks. Within this framework, we first derive the optimal hedging policy for a benchmark loss-neutral firm aiming to maximize expected profit. Our analysis indicates that such a firm might adopt a purely speculative strategy, forgoing forward contracts entirely, when spot market conditions are favorable. Subsequently, to explore the ramifications of managerial behavior, we introduce loss aversion to model the decision-making of a manager who is psychologically sensitive to the heavy penalties for non-compliance. Our analysis reveals that a manager’s inherent loss aversion induces a significant decision bias, causing them to hedge a larger or smaller quantity of allowances than their loss-neutral counterpart. Furthermore, we uncover a pivotal trade-off confronting the firm: a manager’s attempt to maximize their personal loss-aversion utility systematically leads to diminished corporate profits as their aversion to loss intensifies. This finding highlights a critical principal-agent conflict in corporate carbon strategy, offering crucial insights for managing behavioral risks in the pursuit of sustainable operations.
In contemporary global supply chains, geopolitical challenges have become a critical factor contributing to supply chain risk. However, few studies have systematically examined the impacts of geopolitical disruptions from the perspectives of technology innovation momentum and profit loss. To address this research gap, we develop a game-theoretical model of a technology supply chain comprising advanced and inferior technologies, as well as high- and low-quality manufacturers. In practice, geopolitical disruptions – such as the U.S.-China and Japan-South Korea trade wars – are often initiated to undermine the competitiveness of rival economies. However, their actual effects may deviate from initial expectations. Specifically, from the perspective of innovation momentum, we find that geopolitical disruptions reduce the marginal benefit of technology improvement for the advanced-technology supplier, whereas they increase it for the inferior-technology supplier. Moreover, if the advanced technology’s competitive advantage is not significant, the high-quality manufacturer’s innovation momentum will surpass that of the advanced technology supplier. From the perspective of profit, our analysis reveals that while geopolitical disruptions negatively impact the profitability of affected firms, they generally reinforce the profit advantage of the advanced-technology provider. However, one interesting result is that when the technology gap is large, disruptions force the high-quality manufacturer to exit the market, and the advanced-technology supplier incurs a greater profit loss. This is because the high-quality product simultaneously reduces both market share and marginal benefit for the advanced-technology supplier, diverging significantly from the undisrupted scenario.
This paper explores how a strategic marketplace monetizes its information disclosure service and its impact on brands’ competition and social welfare. We conduct an equilibrium analysis of duopoly brands in a horizontal market and a vertical market, and compare the results to those obtained when the disclosure service is free. Under the horizontal market, our results show that costly disclosure will “weaken” one brand by keeping it out of the marketplace, creating a strong incentive for the other brand to invest in disclosure service, whereas free disclosure always lets both brands showcase their products. Interestingly, under costly disclosure, the brand that does not showcase its product can also benefit from its rival’s disclosure. Thus, although the fee is solely paid by the rival, such a cost is “partially” shared by the other brand due to the softened competition. In addition, our results show that free disclosure always reduces consumer surplus and social welfare but costly disclosure decreases the chance that consumers leave the market without buying any brand, and thus reduces consumer surplus to a less extent and might create an “All-Win” situation (except consumers). We also study a vertical market and show some new results: when disclosure service is free, the low-quality brand never solely showcases its product and there exists an “All-Win” area in which all agents (brands, the marketplace, consumers, and social planner) benefit. However, when disclosure is costly, such an “All-Win” area may not exist. Managerial implications are discussed.
Accurate repeat purchase time prediction, i.e., predicting when a customer will repurchase the same consumable product, not only assists businesses and platforms in better understanding their customers but also provides advice for precision marketing. Most prior studies analyze the customer’s repeat purchase pattern from a single view of historical repeat purchase time series, potentially leading to suboptimal prediction performance. However, we contend that the customer’s repeat purchase pattern exhibits both periodicity and uncertainty: the periodicity is driven by the customer’s consumption habit of the product, and the uncertainty arises from the impact of marketing stimulus on the product. Therefore, this paper proposes a Multi-view Information Fusion deep model for Repeat purchase Time prediction (MIF-RT) which simultaneously considers valuable information from the customer, product, and marketing views. Extensive experiments on three real-world datasets have validated the effectiveness of the model compared with other comparative methods.
Effective management of perishable inventories requires balancing uncertain demand, product deterioration, financial constraints, and sustainability considerations. This study develops a unified stochastic inventory framework for perishable goods, where demand follows a Brownian motion process modeled through stochastic differential equations (SDEs). The model simultaneously optimizes the order-up-to level, replenishment cycle, and investment in preservation technology, while incorporating time-dependent deterioration, partial backlogging, hybrid advance-delay payment schemes, and carbon taxation. Closed-form expressions for inventory dynamics across freshness, deterioration, and shortage phases are derived using Itô calculus, and optimal policies are identified through a simulation-optimization approach combining Monte Carlo methods with a two-stage grid search. Numerical experiments demonstrate that while the optimal decision variables under stochastic demand remain consistent with deterministic cases, the average total cost increases by 11
