Economists and perhaps others will benefit from seeing the results of this large-scale randomized controlled trial on wearing masks, in Bangledesh, which studied N=342,126 adults with three study arms: cluster randomized Villages and households with no intervention, with free cloth masks, and with free surgical masks. Participants also received information and local reminders.
Here is the Yale research paper by economist Jason Abaluck et al describing the experiment.
Here is a popular summary in the Atlantic from September 4, 2021
The study shows the superiority of surgical masks over cloth masks, and the both sets of masks achieved a roughly ten percent reduction in symptomatic seroprevalence. This is less than perfect, but the study interventions were only able to increase the wearing of masks from 13% to 42% and raise social distancing from 24% to 29%. Recall that the study was done in Bangladesh.