Severity of COVID-19 Infection and the Adoption of Contact-Confirming Application in Japan

This paper aims to analyze how the prevalence of COVID-19 infection has affected the adoption of the contact-confirming application in Japan and how network externalities have emerged in the diffusion process. Japan's COVID-19 contact-confirming application (COCOA), launched on June 19, 2020, was expected to be a digital platform to work against COVID-19 by notifying users of contact with a positive person, thereby encouraging them to change their behavior. However, the utilization of the app has not been advanced as expected. Given the app's purpose, the number of downloads must be influenced by the spread of infection. A synergetic effect caused by user interdependencies is expected to work for further utilization. In this sense, the effectiveness of the app should depend on the number of users and is therefore affected by network externalities. In addition, registration by positive patients is inevitable for the app to work. In Japan, it is left to the discretion of each infected person to register positive in the app. The number of positive registrations can represent how people are cooperative with the app and thus the Government policy. It would also be interesting to analyze how the spread of infection affects positive registrations. This paper investigates the influence of indices representing the seriousness of the disease on the number of app downloads and positive registrations. Data was collected from the open data published by the Ministry of Health, Labour and Welfare (MHLW). (...)