To me that above comment sounds like rationalizing to justify misleading the public and risking their lives, out of some paternalistic assumption which has not ever been even suggested by the applicable science.
In practice, what little I've seen of people wearing masks causes a change in behavior -- additional distance, etc. from people around them. It raises awareness, and looks like it should reduce risk.
Most of the concern that people might take on additional risky actions would be adequately addressed by making it clear that masks aren't a sure thing-- especially if not expertly fitted and coupled with additional PPE and procedures.
Plenty of people are in "essential" jobs or just need to get groceries. In the bay area, many people already have used masks due to the history of fires... and many of those people are failing to use them because they have been materially and intentionally mislead by people who have, or ought to have, a professional obligation to provide honest information to them.
The loss of trust from this, as the public realizes it was mislead, will have lasting negative effects.
Right now, the average person probably only has 1 or at most a few masks, they don't dispose of them after use and probably wear (put on/off) them in a wrong way.
I'm mostly arguing the other side here, but I should point out that well fitting an N95 mask is actually pretty tricky depending on the shape of your face.
I expect that even a lot of medical professionals don't manage it. You only really can tell that you've done it wrong if exposed to something you can smell or taste very strongly that the mask should stop.
A poorly fitted mask is almost certainly a big improvement over nothing, however.
We'll tell you how many errors you've made.
The fact is that in "the wild", so many different factors are at play that it's very hard to establish the impact of one particular factor on the numbers.
In a sense, establishing the exact impact of wearing or not wearing masks is as difficult as establishing the exact impact of the population pyramid on how things evolve. Combine both and things become even more obtuse.
Worst case is that one ends up confounding correlation with causation which leads to harmful conclusions. Such as over reliance.
What you are saying is the rational sensible middle ground in this discussion. Most domain experts are very careful in the type of recommendations they make.
I'm not a domain expert, and so I put my trust in domain experts. I haven't heard any micro biologist or health care expert urge wearing masks as a basic necessity. I do hear them urge to vigorously wash hands, practice physical distancing and generally stay at home as much as possible...
I used to be strongly anti mask, and I'm slowly changing that position, but I do struggle when I see 60,000 people turn up to horse racing in Cheltenham UK and some of them are wearing masks but doing dumb stuff -- pulling them off to cough or smoke or drink.
* The single most effective intervention we have right now is staying home and avoiding every unnecessary contact, as authorities recommend.
* Worn carefully, several kinds of masks probably have a range of marginal positive effects.
* BUT recommending them runs the risk of (a) people over-relying on them and not following isolation protocols (b) people wearing them incorrectly for a mix of less effective benefits and added risks (c) contributing to a run on masks which constrains supply for medical workers where the benefits are more critical.
If this is all true, it can simultaneously be the case that official guidelines against masks are reasonable recommendations and that it's also a reasonable recommendation to wear masks (on top of other more important precautions).
I'm sure that masks are not the sole factor in gentler curves in those countries. They do seem to be a common factor.