Should we pay people to get vaccinated? Well, maybe.

This past week, a major US retail company, Dollar General, announced that it would pay its employees the equivalent of four hours’ salary if they would get the Covid-19 vaccine. That’s about $40, based on the average pay at Dollar General. The idea is to give employees an additional incentive, and also to cover the time they might need to take off work to get vaccinated.

This is an excellent idea. Let me explain why.

First, though, I should point out that several prominent economists, including Harvard’s Gregory Mankiw and the Brooking Institution’s Robert Litan, have already proposed paying people to be vaccinated–but their proposal is, frankly, terrible. So let’s start with that.

Now that we have two vaccines, from Moderna and Pfizer/BioNTech, with more on the way, we can finally see an end to this awful pandemic. At the moment we have a supply problem: there aren’t enough vaccines to go around. But soon, perhaps in a few months, we’ll have plenty of vaccines. Then the problem becomes getting enough people vaccinated to create “herd immunity.”

(Aside: herd immunity has been discussed ad nauseum this year, so I won’t get into any details, but it essentially refers to the situation where so many people are immune to the virus that it doesn’t spread any more. We probably need 60% of the population to be immune (estimates vary) in order to reach herd immunity.)

A well-informed person might think this won’t be a problem: billions of people are desperate to get the vaccine right now. But the anti-vaccination movement has been spreading misinformation about Covid-19 vaccines since the beginning of the pandemic, long before we even had a vaccine. (Yes, I know it’s patently ridiculous to make claims about a non-existent vaccine, but they did, aplenty. I’m not linking to any of their claims here because I don’t want to give them the traffic.)

As a result of the relentless anti-vax propaganda campaign, a substantial portion of the population is at least “vaccine hesitant,” meaning they’re not sure if they want the vaccine. They are worried primarily about safety, even though the data is very clear that these vaccines are remarkably safe. (It’s true that a tiny number of people have had allergic reactions, but this data is public and no one’s hiding it.)

So we need to convince some people that it’s in their own best interests to get vaccinated. A small number of deeply confused anti-vaxxers, such as the people behind the mis-named NVIC, are probably unreachable. They simply won’t listen, preferring to believe their own misinformation and conspiracy theories. But for the large number of people who are merely hesitant, a positive incentive might be just the thing to convince them to get vaccinated.

Enter the economists. Robert Litan first proposed paying people to take the vaccine back in August, and Gregory Mankiw strongly endorsed the idea, writing:

“what’s the best way to achieve herd immunity? Again, simple: Once a vaccine is approved, pay people to take it.”

Such confidence! Actually it’s quite a good idea in principle. But Litan and Mankiw then went off the rails, proposing that we pay everyone $1000 each to get the shot. Litan admitted that he didn’t have any data to support this particular amount, but he called it a “strong hunch.”

That’s a $300 billion program. Neither Litan nor Mankiw was bothered by this.

There are some gigantic problems with this proposal. First, because it’s such an enormous amount, it’s extremely unlikely that it will ever happen. It’s just the kind of hypothetical, pie-in-the-sky proposal that gives academics a bad name. Because it will never happen, Litan and Mankiw will never have their idea tested in real life, and no doubt they will continue to claim it would have worked.

Second, though, is a much bigger problem, as pointed out by economists George Loewenstein and Cynthia Cryder in the New York Times, and by medical ethicists Emily Largent and Franklin Miller in JAMA. The problem is that if you offer to pay a lot of money to do something, then people conclude “this is something you would not want to do without compensation.” In other words, it’s dangerous or somehow bad.

Thus a large payment may merely heighten people’s suspicions that the government (or “Big Pharma”) is up to no good, and that’s why they have to bribe people to take their suggestions. The last thing we need right now is to increase people’s mistrust of vaccines.

In addition, paying so much money for each shot is, as Largent and Miller point out, a bad investment. Sure, $300 billion is much less than Covid-19 is costing us right now, but it’s still a huge sum, and those funds could be better spent on many other things, such as helping to support states that are still struggling to set up facilities to administer the vaccine.

Now let’s go back to the Dollar General plan. Dollar General is paying the equivalent of four hours’ worth of salary, about $40 on average, to each employee who gets vaccinated. As I said above, this is an excellent idea.

Why is this better than the $1000 per person plan from the economists? First of all, it costs far, far less than the economists’ plan, which makes it far more likely to happen. Second, Dollar General is paying people to defray the actual costs–in time–that they will incur in order to get the vaccine. So it’s not so much a bribe as it is a modest reimbursement. Third, by providing a modest payment, they provide a relatively bigger incentive to low-income groups, and they avoid paying billions of dollars to high-income people who are already highly motivated to get the vaccine.

I suggest we adopt a version of Dollar General’s plan for the entire U.S. population. Why not offer a cash payment of $20 to everyone who gets the vaccine, and pay it immediately? Obviously there would be some logistical challenges to this–we’d need security procedures to make sure the $20 payments were actually handed out and not stolen–but it would be far simpler than arranging the $1000 payments so blithely proposed by the academic economists. And it would only cost $6 billion rather than $300 billion.

A payment of $20 provides a small positive incentive, and it can be justified as paying people for the time they spend getting the shot. Because it won’t seem like a bribe, it will be much less likely to raise suspicions that the vaccine is harmful (which it isn’t, I hasten to add).

So Litan and Mankiw were sort of right: paying people might encourage more people to get vaccinated. But they’re wildly wrong about the size of the payment, which other economists and bioethicists have pointed out would likely create distrust. Dollar General is getting it right: let’s offer everyone a small cash payment if they’ll take the time to get the vaccine. It might just work.