I’m now fully vaccinated.  How should I change my behavior?  How should anyone?

One popular answer is: Not at all.  Why not?  The top reason I’ve heard is: Because even those of us who have been vaccinated can’t be absolutely sure we won’t be infected – or spread infection to others.  Some use the same reasoning to argue that people who have recovered from COVID shouldn’t change their behavior either.  As immunologist Alexander Sette puts it:

Not taking any precautions—including wearing a face mask, practicing social distancing, or getting vaccinated—after an initial coronavirus infection is comparable to “driving a car where you’re 90% sure the car has brakes.”

However, both common sense and economic reasoning say virtually the opposite.  If a risk falls by 90% – and there are large gains to accepting the risk – you should not only accept more of the risk; you should probably accept much more risk.

This is obviously what self-interest recommends.  And when your risk-taking benefits others, this is what humanitarianism recommends as well.  Remember: Your social distancing doesn’t just harm your quality of life. Your social distancing also harms the quality of life of everyone who loses the pleasure of your company and the profit of your patronage.  (Caveat: Since vaccines take two weeks or so to kick in, neither self-interest nor humanitarianism recommend drastically changing your behavior the instant you get vaccinated).

What about the “90% sure the car has brakes” analogy?  It posits an lopsided scenario where you have a 10% chance of killing or seriously injuring others for a trivial total benefit.  You shouldn’t die with 100% probability to see a movie; neither should you die with a 10% probability to see a movie.  Anyone who has ever driven to a movie, however, has accepted a .00001% chance of dying en route.  And accepting such a risk to see a movie is both prudent and considerate.

Or to tweak the hypothetical, it would be perfectly reasonable to drive regularly even though there is 10% chance that your brakes will go out sometime in the next twenty years of driving.  Modest risk, massive gain.

The better argument against changing your behavior – or at least not changing it much – is that we still don’t know if vaccinated people are contagious.  The absurd yet popular version of this argument is that we can’t be 100% sure that vaccinated people are 100% non-contagious.  Caricature?  Hardly.  Here’s the Mayo clinic on March 5, 2021:

Keep in mind that the Pfizer-BioNTech COVID-19 vaccine is 95% effective in preventing the COVID-19 virus with symptoms. The Moderna COVID-19 vaccine is 94% effective in preventing the COVID-19 virus with symptoms. The Janssen/Johnson & Johnson COVID-19 vaccine is 66% effective at preventing the COVID-19 virus with symptoms. While your risk of getting the COVID-19 virus after being vaccinated is low, it is possible.

It’s also not clear if the COVID-19 vaccines reduce the spread of the COVID-19 virus. As a result, it’s not known if a person who is vaccinated could be a carrier of the COVID-19 virus and spread it to others, even if he or she doesn’t become sick. More research is needed to determine if you are still contagious after being vaccinated.

Because of these factors, even once you’re vaccinated you could still pose a health risk to unvaccinated family and friends by visiting with them in person.

Fortunately, there is a simple way to repair this argument.  Namely: Hold reasonable beliefs about how contagious you are – and act on those beliefs.  How does one acquire these “reasonable beliefs”?  Textbook Bayesian inference.

Step 1: Get a reasonable base rate.  How much does the typical vaccine reduce contagion?

Step 2: Adjust that base rate in response to evolving evidence.

Step 1 is pretty simple.  Vaccines have an excellent historical track record for not only protecting the vaccinated, but virtually eradicating a long list of contagious diseases.  As Paul Sax puts it:

So far, the single question that has drawn the most attention is this one:

Do the vaccines prevent transmission of the virus to others?

In my response, I try to highlight the fact that while we don’t have ironclad proof, it is highly likely that they will lower the risk. I urge you to read the full question where I outline the evidence so far.

In my response, however, I wrote this:

If there is an example of a vaccine in widespread clinical use that has this selective effect — prevents disease but not infection — I can’t think of one!

Some colleagues now have pointed out a few examples — diphtheria, meningitis B, and pertussis. My apologies for not mentioning these! We will update the site, and thanks for pointing this out.

Nonetheless, the general (if not ironclad) rule that vaccines typically reduce the risk of transmission to others remains true.

Bottom line: In the absence of any further evidence, we should have high confidence that COVID vaccines, like the large majority of vaccines, greatly reduce contagion.

Step 2 is harder.  There is favorable preliminary evidence (scroll to Do the vaccines prevent transmission of the virus to others”) but as a rule I’m not a fan of “preliminary evidence.”

The right reaction to this situation, though, is not to be agnostic, but to rely heavily on Step 1.  The burden of proof is on those who doubt the contagion benefits of vaccines – and as far as I can tell, there is no good reason for doubt.  Yes, we should further update our beliefs as new evidence comes in.  But for the time being, we should act on the belief that vaccines sharply reduce risk to others as well as to ourselves.  This is what I am doing myself – and what I advise you to do as well.

Note: Relying on base rates when specific evidence is scarce is distinct from the libertarian presumption against coercion.  The libertarian presumption says governments shouldn’t force you to take extra care unless doing so yields large social gains with high certainty.  What I’m saying, in contrast, is that we should already be highly certain that forcing vaccinated people to take extra care will yield large social losses.

I suspect that many people are so dehumanized by the COVID experience that they don’t consider a few more months of hyper-caution to be a large social loss.  Yet if you do the math, you’ll see that the combined harm of private and government COVID prevention destroys millions of life-years per month in the United States alone.  The mass loneliness, isolation, and fear we’ve endured stagger the imagination.  Every vaccinated individual who returns to normalcy is doing themselves and the people around them a big favor.

Remember: The sociable vaccinated also help the unvaccinated escape loneliness, isolation, and fear.  Tonight is a great time for the vaccinated to invite their unvaccinated friends over for dinner and boardgames.

The main roadblock is that other people can’t easily tell if you’ve been vaccinated.  Your close friends will know, but what about everyone else?  Hence, when you return to normalcy, you might just frighten others instead of putting them at ease.  In a sensible world, we’d have brightly-colored vaccination bracelets to spread goodwill.

Since that isn’t going to happen, however, I’ve decided to fill the void.  A couple weeks ago, I launched an illustration contest for COVID vaccination wear.  I told entrants to combine the words, “Fear me not! I got my COVID vaccine” with artwork expressing friendship and joy.  There were almost a thousand entries, and I selected three winners.

Long story short: Starting today, you can order t-shirts, sweatshirts, jackets, and more from my new Zazzle store, FearMeNot, to let the world know you’ve been vaccinated and are ready to return to normalcy.  I told the artists to go for pure joy, and they delivered.  Check out the winners:

So let the rehumanization of humanity begin, one vaccination shirt at a time.  If you order now, you get 15% off with the coupon code MONDAYTREATZ.  And check out FearMeNot store for other way to share your return to normalcy with the world.

P.S. No, I am not worried that more than a tiny fraction of unvaccinated people will falsely wear my shirt.  That is paranoia talking.