Ohio Gov. Mike DeWine’s announcement last week that his state would use a lottery to persuade more people to get vaccinated breaks new ground in the struggle to reach the vaccine hesitant. Starting May 26, five people will take home $1 million over five consecutive weeks; anyone with at least one shot will be eligible. What’s more, five vaccinated Ohioans under the age of 18 will get a four-year college scholarship, doled out in a parallel drawing.

DeWine (R) is eager to open his state — he’s dropping almost all restrictions on June 2 — and has settled on a psychologically interesting approach to get shots into arms, one that is likely to have more of an impact than programs that are offering free drinks to vaccinated residents, or West Virginia’s $100 gift cards or savings bonds for people ages 18 to 34.

Some of the attractions of lotteries are familiar — who wouldn’t be intrigued by the possibility of winning $1 million? — but they also take advantage of human psychological quirks. Lotteries are tremendously popular because people tend to focus on the magnitude of the awards and overestimate the small probability that they will win — near zero, in some cases. Since about 5 million people have been vaccinated in Ohio at this point, the odds of winning are already quite small (roughly 1 in a million over the five weeks — with the chances going down as more people get shots). Still, it is likely that by now nearly everyone in Ohio knows the size of the reward that’s being handed out.

People don’t balk at spending money with a minuscule chance to win large prizes, so it seems plausible that people on the fence about vaccines will be motivated by the new program to change their minds. A concept called “anticipated regret” also plays a part in encouraging Ohioans to get shots. People who have failed to get vaccinated will feel a rueful twinge when they see the winners announced. Ohio is ramping up this effect by drawing their winners not from a list of vaccinated people but from the state’s voter registration database: Ideally, there would a drawing and people would be notified if they would have won had they been vaccinated, but had to be passed over. (Taking this thought experiment to an extreme, the state could even publicize the names of those who would have won $1 million had they been vaccinated.) Such psychological effects aside, there is also the entertainment value of the lottery, as undoubtedly there will be lots of news coverage of the winners, their life stories and so on — magnifying the publicity for the outreach effort.

While the million-dollar prizes are getting a lot of attention, the cost per new person vaccinated may turn out to be quite reasonable. Suppose 1 million more people out of Ohio’s total population of 11.6 million get shots: The cost per incremental new vaccinated person will have been roughly $5, much less than a $100 gift card. And with a lottery, as the number vaccinated goes up the cost per person goes down. This makes the cost structure significantly different from per-person payouts.

While financial incentives have proved effective in increasing healthy behaviors ranging from smoking cessation to physical activity, it is difficult to estimate the effectiveness of a specific incentive amount or design — especially in the present context, given that many Americans have already been vaccinated and the myriad political and social factors influencing vaccination decisions. A randomized survey of more than 14,000 unvaccinated people by the UCLA Covid-19 Health and Politics Project, for example, suggested that even small payments would make a difference: 34 percent of respondents said they’d be more likely to get vaccinated if they were paid $100; a similar proportion, 28 percent, said that they’d be more likely to do so for a mere $25. With the experience in the states experimenting with payments, we’ll soon have a better sense of real-world effects.

There are a number of concerns with offering a financial incentive for vaccination. One is that financial incentives offered now could engender expectations about receiving financial incentives in the future should booster shots be needed: Payments beget payments. A time-limited lottery, however, may engender fewer expectations than a one-for-one incentive for vaccination.

People who have already been vaccinated could be upset that those who dragged their feet are now being awarded; the Ohio program avoids that by making all who have been vaccinated eligible for the lotteries. Another, more theoretical concern is that offering an incentive could signal to people that the vaccination is undesirable or unsafe; in other words, the government is paying you to compensate for a risk. But now that more than 100 million Americans have voluntarily been vaccinated, that is less likely to be a problem.

Ohio is undertaking a fascinating experiment. The concept of using lotteries with winnings of the magnitude being used here to motivate behavioral change in large populations is a genuinely fresh approach.

Some people find the idea of providing financial incentives for health behavior objectionable, on principle. But given the substantial public resistance to public and private vaccine mandates, as well as to “vaccine passports” and similar approaches that could be highly effective, a program of this kind could turn out to be a cost effective way to reduce future health and economic costs from covid-19. Given the stakes, Ohio’s approach is well worth testing.