Subsidy Strategies for Drone Food Delivery under Dual Network Externalities: A Four-Party Stackelberg Game Approach

Imagine a city where the air is filled not just with birds and planes, but with small, buzzing drones carrying hot meals to hungry people. This vision promises to solve real problems: it could clear our streets of the traffic jams caused by delivery drivers, reduce pollution, and make food arrive faster. Yet, for all this potential, the widespread use of drone delivery remains stuck in a holding pattern. The technology exists, but the market is hesitant. The infrastructure, like the special landing pads and charging stations needed for these machines, is sparse and unreliable. At the same time, the cost to the consumer is often too high, and many people feel a deep-seated unease about letting a machine handle their food. To get this system moving, governments are considering stepping in with financial help, but the question is tricky: who should get the money, and how much?

A team of researchers from universities in China has tackled this puzzle by building a complex mathematical model to simulate how different subsidy strategies play out. They did not just look at the price of a meal; they constructed a dynamic scenario involving four distinct groups: the government, the companies building the drone infrastructure, the food delivery platforms that manage the orders, and the customers themselves. The researchers wanted to understand how these groups interact when two powerful social forces are at work. The first is a kind of "network effect," where the service gets better and more valuable for everyone as more landing spots are built. The second is a "peer effect," where people are more likely to try a new service if they see their friends and neighbors using it. However, these positive forces are often held back by "psychological costs"—the invisible mental barriers of fear, privacy concerns, or simply the worry that the technology might fail.

The study, which uses a method of working backward from the final outcome to determine the best starting move, tested three different paths for government intervention. In the first scenario, the government offered no help at all, serving as a baseline. In the second, the government gave money directly to the companies building the landing pads and charging stations. In the third, the government gave money directly to the customers to lower the price of a drone-delivered meal. By running these simulations, the researchers discovered that there is no single "best" way to subsidize the industry; the right choice depends entirely on the specific conditions of the market at that moment.

The most striking finding is that the effectiveness of peer pressure—the idea that people follow the crowd—is not always positive. The researchers found that this social influence only helps the drone industry grow if the psychological fears of the customers are low enough. If people are deeply worried about privacy or safety, seeing others use the service might actually make them more hesitant, not less. In this high-fear environment, the "crowd" effect can backfire. Furthermore, the study revealed that the strength of the infrastructure network changes the rules of the game. When the network of landing spots is weak, the government must decide carefully whether to build the physical foundation or to encourage the customers.

The simulations showed that if the network of infrastructure is weak and the psychological fears are low, but the "crowd" effect is also weak, the government should subsidize the customers to jumpstart demand. However, if the infrastructure is weak but the "crowd" effect is strong, it is better to subsidize the companies building the infrastructure, as the social momentum will naturally pull customers in once the service is available. As the network of infrastructure grows stronger and reaches a certain level of maturity, the market begins to function on its own. At this point, the researchers suggest that the government should begin to withdraw its financial support in an orderly fashion, allowing the market to sustain itself without artificial aid.

Ultimately, the paper argues that a one-size-fits-all policy will fail. The optimal strategy is a shifting one that evolves as the market matures. In the early stages, when the network is thin and fears are high, the government needs to be a careful architect, choosing between building the physical roads or paying the travelers. Once the network is robust and the technology is trusted, the government's role should fade, letting the natural forces of supply and demand take over. This approach offers a clear roadmap for policymakers who want to turn the promise of drone delivery into a reality without wasting public funds on strategies that do not fit the current stage of development.