Early in the Covid-19 crisis, the wearing of masks in the US became a political statement. I'm sure everybody's thinking about why this happened is fairly well cemented, so rehashing how the US got there is not really helpful.
The anti-mask movement has spread from the US, where it is present among a large number of people with a disdain for science, to people in other countries. I am a plaintiff in a lawsuit in British Columbia, Canada trying to get the schools to be able to reopen in a safe manner. The right thing to do, from an epidemiological perspective, is to require masks. This is no different than requiring seat belts, mandating food safety in restaurants, or prohibiting drunk driving. There is actually a far stronger case for masks -- they save lives -- than there is for laws prohibiting public nudity or requiring a permit to use public spaces.
We live in the world we have, not the world we wish we had. In the world we have, a substantial number of people refuse to take the basic steps to keep us safe during a pandemic. A substantial number of people even profess to believe that those public safety steps actually make you sick -- akin to claiming that a condom increases your risk of catching an STI. But we cannot let public health perfection stand in the way of better public health.
Governments and private entities are acting based on the incorrect assumption that masks are a binary choice: Either every place requires them or every place does not. There is another way. Granted, this method will not stop the virus from spreading, and case numbers may keep going up. But they would mostly go up among populations that choose to take the risk (and the unfortunate relatives of that population who are exposed at home).
Taking schools as an example, the lawsuit I am involved with takes place against a background of the British Columbia school reopening plan. That plan splits schools into learning groups or cohorts -- rather large ones at that, reaching up to 120 students in each cohort for high schools and 60 students in primary schools. A family like mine is thus exposed to the risk of catching Covid-19 from any of 180 students (one high school student and one primary school student).
Now imagine if instead of cohorts that are identical, cohorts were instead designated "high risk, high safety" and "low safety". Parents would elect to put their children into one of the two groups. High risk teachers and administrators would be able to elect the high safety cohort to work with, while teachers required to work with the low safety cohort would be given hazard pay (given the risk to families, I'd imagine the hazard pay would have to be substantial to get enough teachers willing to take the risk).
The high safety cohort would be kept entirely separate from the low safety cohort. Surfaces would be sterilized after being exposed to low safety cohort members. Low safety and high safety cohorts would be completely separated, operating on different schedules. Perhaps critically, because the high safety cohort has elected to be safe, there would be near total compliance with health recommendations. No amount of anti-mask propaganda from parents would matter, because the anti-mask parents would all elect the low safety cohort.
In the case of British Columbia, this would mean that one cohort would use the existing rules, and another would follow much safer rules. Elsewhere, it may mean that non-compliance disappears where the existing rules are already high safety.
Turning to retail stores, a similar program could be put in place. Entire stores could elect high safety or low safety. The stores would have to make reasonable accommodations for those employees who are at high risk (per the US ADA and Canadian human rights law), perhaps letting high risk employees at low safety stores work in the back room. In some cases, there may be two Starbucks a one minute walk away from each other (this is a real thing). One Starbucks may be high safety and require masks, have no indoor seating, clean high touch surfaces every ten minutes, etc. The other Starbucks may be low safety and not require masks, have indoor seating, and be more lax in cleaning surfaces. While hazard pay would likely be required to get employees to take this risk (or they could find enough anti-maskers among their ranks and do inter-store transfers), it would create a situation where people who are at risk can actually go to a Starbucks without worrying about non-compliance with safety rules.
In places where shopping density is too low to have some stores high safety and some stores low safety, they could in theory split the safety level on a time basis. For example, every odd numbered hour would be low safety, every even numbered hour high safety, with a surface sterilization every time there is a switch from low to high safety.
On an economic basis, this is likely to encourage people who have not been willing to risk shopping (I haven't seen the inside of a supermarket in six months) to shop. It will dramatically reduce the drop in consumer spending by allowing those of us at too high a risk level to actually get out and spend the money we have. In other words, we could add a point or two to the Gross Domestic Product. A similar economic effect would happen with the school changes. Parents who are too high risk to allow their children to risk bringing home the virus are currently faced with lower productivity as they have to split their time between home schooling/childcare and work. If the schools provide a safe alternative, those parents would be able to work more -- again improving the GDP.
From an epidemiological perspective, the perfect approach would be to lock down for several weeks and basically eliminate any endemic cases. But the political will to do so does not exist. A less effective approach, but still a good deal better than nothing, is to free those of us at high risk to participate equally in society. If somebody was randomly shooting people, it would obviously dampen the economy as people would not want to be exposed. For those of us at high risk, the dampening effect of Covid non-compliance is very similar. This middle ground, catering to science believers and science deniers, allows society to reopen more safely while reducing the economic impact of the virus.