Source: Johannes S. Kunz, Carol Propper, and Trong-Anh Trinh, “The Impact of Internet Access on COVID-19 Spread in Indonesia,” Asian Development Bank (ADB), ADB Economics Working Paper Series, no. 723 (April 2024).
Commentary: A recent working paper by Johannes S. Kunz, Carol Propper, and Trong-Anh Trinh analyzed the role of broadband internet access on the spread of COVID-19 in Indonesia. Using data from the Indonesian COVID-19 Task Force, the authors looked at data for 416 regencies (administrative divisions) and 98 cities between March 2020 and February 2023. Data on mobile broadband coverage came from Collins Bartholomew’s Mobile Coverage Explorer. Specifically, the authors used 3G coverage as their main proxy for internet access. Additionally, the authors controlled for a variety of demographics factors including, but not limited to, population density, the share of the population over age 65, and the average household size. The study’s findings suggest that mobile broadband access had a significant effect in reducing COVID-19 cases.
The authors found that, overall, 3G internet access was associated with a 25 percent decrease in COVID-19 cases. They noted that 3G internet access had at least as significant an impact on COVID 19 cases as other policies such as masking and social distancing. They also found that 4G access was associated with a similar reduction in COVID-19 cases. The authors also established causation using an instrumental variable (IV) approach, using frequency of lightning strikes as their instrument. They argued that lightning strikes could exogenously affect internet access, as such strikes can often result in loss of internet connection. The results of their IV approach suggested that internet access resulted in a 25 percent to 50 percent reduction in COVID-19 cases.
The authors noted a few areas where there was variation in the benefit of internet access. For instance, they noted that internet access reduced COVID-19 cases to a less extent in rural areas than urban areas. They attributed this to the fact that the types of economic activities dominant in rural areas, such as agricultural work, do not have the option of remote work. They also mentioned that it may also be attributable to lower rates of digital literacy in rural areas. At the same time, they found that internet access was more effective at reducing COVID-19 cases in areas that have a greater share of workers with remote work options.
