The effectiveness of mandates.

Breathing in smoke is bad, mmmkay?

But that’s not the question.

Forest fires and n95 masking:

The question is: when you live in a city blanketed with smoke, should you advise citizens to mask when they walk outside, etc? Let’s think through that.

First, recognize that some baseline exposure will happen no matter what. Even in your house, you still can smell smoked salmon. The baseline exposure may dwarf any portion remotely modifiable.

Second, recognize that a tiny fraction of people will mask anyway, and others will exercise outdoors without a mask bc they don’t care. Ergo, your policy will at best affect a tiny portion of people. Those who will now mask who otherwise wouldn’t. This may be a vanishingly small group. Yet, the success of the policy hinges on this!

Now think it through a little more.

How much will they mask? Can they sustain it? Will they have gaps at their nose? Will they alter their behavior (go out more when they otherwise wouldn’t?) Aka the Peltzman effect?

In essence, the societal benefit of a masking requirement is not the same as the individual benefit of masking.

Specifically, a masking requirement will not change the minds of those who are already preparing to mask for their health, and those who are refusing to mask (or are forgetting to mask) probably won’t change their minds.

On the margins, you’re only altering the behavior of a small subset of people–and chances are very good they won’t persist in masking anyway. (How many chin-diapers did you see during the pandemic? How many exposed noses?)

In exchange for this requirement whose benefits are marginal at best (as a requirement is only as good as the behavior it alters and the benefits of that altered behavior), you spend hard-earned social capital that cannot be easily regained in the future. Social capital that cannot be spent on other tasks.

And when you consider this requirement is about changing behavior, you have to consider other potential paths towards changing behavior. Such as an educational public service announcement which briefly explains the benefits, rather than a mandate people that people increasingly ignore.


Worse, you cannot simply pass a policy then, when people change their behavior, you can’t claim a victory: you don’t know–because you haven’t conducted the randomized trial–if the people who choose to wear masks would have put one on anyways. It’d be like passing a mandate about wearing jackets when it’s snowing outside: you can’t declare victory because people then put on jackets before going outside.


Finally, the real motivation for me writing this column is that public health is being led by the dumbest people. They want us to accept that bioplausibility is the basis for costly, recurring policy choices and it simply isn’t. You need real world, randomized data.

Published by