The government intervention effects on panic buying behavior based on online comment data mining: a case study of COVID-19 in Hubei Province, China

At the end of 2019, the world grappled with an unparalleled public health crisis due to the COVID-19 pandemic, which also precipitated a global economic downturn. Concurrently, material panic buying occurred frequently. To restore benign market order, the government instituted a series of interventions to stabilize the market. This scholarly exploration dives deep into evaluating the tangible impact of these governmental measures in Hubei, China, a region which found itself at the very epicenter of the epidemic in its onset phase. Existing papers often employ structured questionnaires and structural equation methods, with small samples and limited effective information. In contrast, we used a dataset of tens of thousands of entries and employed text analysis to maximize the extraction of valid information. Through a meticulous analysis of public feedback, our findings unveil several pivotal insights: (1) The news measures of materials have the best effect. Their effectiveness, in descending order, is ranked as: material sufficiency > authority effect > market supervision > appeal and guidance; (2) Government measures during the epidemic’s initial phase exhibited a delay. After the lockdown measures, the phenomenon of large-scale buying has been formed, and the relevant material news was released later; (3) A dual approach combining authority influence with material sufficiency yielded the most favorable results. In light of these findings, the paper concludes with tailored recommendations aimed at amplifying the efficacy of government-led public opinion interventions in future crises.