Due to its ability to facilitate high-frequency and high-volume transactions within a short period, live-streaming e-commerce has garnered significant attention from both industry and researchers. By comparing live-streaming e-commerce to the stock market, it is suggested that herd behavior in purchasing may be a critical factor contributing to the high-frequency and high-volume transactions observed in live-streaming e-commerce. Unlike traditional e-commerce, live-streaming e-commerce allows focal consumers to witness the purchasing process and resulting outcomes of preceding consumers in real-time, potentially adding a new dimension to preceding consumers' purchasing behavior that goes beyond product quality information. To ascertain whether herding effect exists in live-streaming e-commerce purchasing decisions and to uncover the true informational content of preceding consumers' purchasing signals, this study collected a dataset comprising 371,829 sales sequences for 42,339 products from Douyin Live, a leading live-streaming e-commerce platform. Utilizing theories of information cascades and scarcity cues, the empirical analysis demonstrates the presence of herding effect in consumer purchasing decisions within live-streaming e-commerce. Furthermore, it reveals that focal consumers primarily interpret the purchasing behavior of preceding consumers as scarcity cues. This research not only theoretically elucidates a key reason behind the rise of live-streaming e-commerce but also clarifies significant differences between live-streaming and traditional e-commerce. These findings offer valuable managerial insights for companies managing live-streaming sales and for platforms regulating live-streaming e-commerce activities.