How to think about risk and travel this Thanksgiving : The Indicator from Planet Money The CDC recommends we all stay home for Thanksgiving to minimize the risk of spreading COVID-19. But tens of millions of Americans are expected to travel anyway. If you're one of them, here's how you can think about the risk you're taking.
NPR logo

Should I Travel For Thanksgiving This Year?

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Should I Travel For Thanksgiving This Year?

Should I Travel For Thanksgiving This Year?

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript




Hey, Allison.


VANEK SMITH: Allison Schrager is an economist with the Manhattan Institute. She specializes in risk.

OK, so I am calling you because you sent me a text. And you said, have you thought about doing a show about Bayesian updating? I know we talked about risk assessment, but going into next year, Bayes - is it Bayes? - seems important to understand. So you don't typically reach out to me in this way. So I - when I got this text, I was like, whatever this is, it's really important. I need to call Allison up and figure out what this is and why we all need to know this (laughter).

SCHRAGER: Yeah, so I thought of it because I was watching TV, and, of course, they were telling you, following CDC guidelines, which I am not disputing at all - even if you get a test, you should not visit relatives for Thanksgiving.

VANEK SMITH: Oh, like, even if you test negative for COVID.

SCHRAGER: Yeah, even if your plan is, I'm going to get a rapid test, then I'm going to go see my family, still don't do it. It's still high risk. And - or it didn't say high risk. It's like - it's still like everyone's going to get sick. They said it almost was a certainty. And it struck me that, you know, when you think about how you're going to manage risk in COVID, which is really a very sort of difficult risk problem, I said it was not very Bayesian.

VANEK SMITH: Not very Bayesian - this is THE INDICATOR FROM PLANET MONEY. I'm Stacey Vanek Smith. Risk - it's pretty much all we hear about lately; the risk of traveling, the risk of not wearing a mask, the risk of spending the holidays with live people, the risk of COVID. Today on the show, economist Allison Schrager walks us through how she's approaching risk this holiday season and how we can better assess our own situations.


VANEK SMITH: Bayesian - so, OK, who - what is Bayesian? It looks like it's this man, Thomas Bayes.

SCHRAGER: Yes, who was a mathematician/philosopher, and he came up with this thing called Bayes' Rule.

VANEK SMITH: In the 1700s, so this was a while ago, but he came up with this rule. And what did the rule say?

SCHRAGER: It was pretty much how you can estimate the probability of one thing happening given that another thing has happened or is likely to happen. So the odds that you're going to go see your grandmother and then you're going to kill her, you know, is one probability. The odds that you're going to go see your grandmother and you're going to kill her given that you have a negative test is a different probability.


SCHRAGER: Neither are zero, but they are different. And that's what I mean by Bayesian. Like, you have it - there's a certain probability you have a test, it's a negative result; so these are the odds you're not going to infect your grandmother.

VANEK SMITH: So how is this coming into play now? - because we're in sort of this particular moment with COVID where the cases are surging, cities, like - it seemed like things had been opening back up. But now it sort of seems like the trend nationally is that they're closing back down. So how does this play into Bayes and risk and how we live our lives and deal with the holidays?

SCHRAGER: Well, the idea is you want to update your probabilities and risk assessments as new data comes in.


SCHRAGER: Right. You know, your risk assessments of what's safe to do are going to change if your community infection rate is 2% versus 50%.


SCHRAGER: Right, so you want to account for that. Also, coming in next year as, say, treatments like antibody treatments become more available - that also might affect your risk assessment, so it's always updating with new information. In fact, if you remember back in March, when they said 2 million Americans were going to die - and there's been all this criticism about those projections. It turns out that some of those projections were right because they didn't account for human behavior that people might distance. So that was un-Bayesian. Bayesian would always be like, all right; given this new information, let's adjust our probability estimates.

VANEK SMITH: It seems like everybody's gone one of two directions; either, like, stay inside with a mask and have no contact with other people, or throw your mask to the wind. Forget it. Let's just live our lives. It seems like there's been sort of a polarization. So is there, like, a Bayesian middle ground?

SCHRAGER: Yeah, if you're a Bayesian, you would occupy the middle ground, which is you'd adjust your behavior as risks change, as information changes, right? Like, back to your grandmother projection, like, how risky is this? Would it change if she got an early vaccine, right?


SCHRAGER: So that would be a Bayesian - like, all right. Well, I mean, a vaccine's not 100% effective. It's 90-, 95% effective or, I guess, 94% effective if you're elderly. So maybe then you would be more open to it. And this is going to be really important in the coming year as the vaccines are rolled out. Some people are going to have it. Some people don't. It will take a while for community herd immunity, so our risk calculations are going to have to be changing really rapidly as next year rolls on.

VANEK SMITH: So what would a Bayesian Thanksgiving look like this year?

SCHRAGER: Well, it depends on your risk tolerance, and it also depends on where you live and how you could feasibly travel.


SCHRAGER: Right? If you're in a really high-breakout state, you know, probably the odds of infecting other people and worsening the spread is very, very high. If you're somewhere where it's less risky and you take a lot of precautions like getting tested, you know, traveling by car - still not a guarantee but probably a lower risk. And as I said, you want to, when you make those risk decisions, incorporate, as I said - like, if I have a negative test, what is this probability? Not zero but certainly lower than if I don't test.

VANEK SMITH: So what about you for your Thanksgiving? What are - I know you live here in New York. What are you - what are your Thanksgiving plans? Are they Bayesian?

SCHRAGER: They are. And I've have struggled with it because, you know, my mother doesn't live very far away. And so I can get there easily. But...

VANEK SMITH: Like, you don't have to get on a plane.

SCHRAGER: No, I can get on a train. I've already gotten one rapid test, and I'm going to get a second one morning of leaving. And I chose to go on a more expensive train that is emptier where you can guarantee your own seat. So I am taking a - I'm taking a Bayesian approach.

VANEK SMITH: OK. And so your mother is - obviously has not gotten a vaccine because nobody really has. She's in a higher risk group maybe. So...


VANEK SMITH: You have, like, sort of tried to mitigate risk in the kind of transportation you're taking. And you're getting a couple of tests.

SCHRAGER: Yes, two tests, one morning of - and again, it's not a guarantee. I'm definitely taking more risk than if I sat at home. But as I said, I'm mitigating the risk to some extent.

VANEK SMITH: How do you think, as a country, we are handling the COVID risk that we're facing right now? I feel like it's a little - like, it's sort of - government-wise, it's such a hodgepodge of, like, different cities and states having different reactions. And I mean, when you're looking at this as an economist who specializes in risk, like, what do you see?

SCHRAGER: Well, I think the biggest failure policy-wise is the lack of testing. But also, I think the other huge public health failure that people talk about a lot less is our risk communication. I'm of the very small school that believes people can make good risk decisions, but they have to have the information presented to them in ways that make sense to them, that's natural to them and that makes sense. And the problem is is you get sort of very mixed messages. You get either, it's fine; go about your life, or, gee, you know, this is terrifying. Never leave your home. And I think that's been the big public health failure that we're not really talking about.

VANEK SMITH: All right. Do you do any other economisty (ph) things for Thanksgiving? Do you, like - I don't know - divide the pie in a special way or something?

SCHRAGER: Oh, yeah, I'm an economist all the time. Even getting testing, it's a three-hour wait where I live right now to get a test in New York. And I got - so I was in line the other day after three hours and got so annoyed because then I learned that there's a private clinic that, for an exorbitant fee, will just test you for money. And there's no wait. So then I'm, like, in line being like, what's the value of my time? And is this worth the fee that I thought was outrageous? So I might splurge and pay for a test.

VANEK SMITH: Instead of taking one of the free...

SCHRAGER: They're free. But, like, I noticed as I was talking to people in line that no one else was thinking about opportunity cost the way I was.

VANEK SMITH: (Laughter).

SCHRAGER: I'm like, this is three or four hours of your time.

VANEK SMITH: Well, Happy Thanksgiving, Allison. Thank you for talking with us.

SCHRAGER: You too.

VANEK SMITH: And I hope you have a great time with your mom.

SCHRAGER: We'll be very Bayesian.


VANEK SMITH: This episode of THE INDICATOR was produced by Nick Fountain, fact-checked by Sean Saldana. THE INDICATOR is edited by Paddy Hirsch and is a production of NPR.


Copyright © 2020 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.