The Washington PostDemocracy Dies in Darkness

Imposing vaccine mandates may be counterproductive, our research suggests

Mandates may increase distrust and public resistance

A man cools off in a fountain in a Manhattan park on May 19, when coronavirus restrictions were lifted in New York. (Spencer Platt/Getty Images)

Opposition to coronavirus vaccination stubbornly persists across the globe. Even in Israel — the world leader in vaccinating its population — the fraction vaccinated appears to have hit a plateau with one-third of the population unvaccinated. In the United States, some states are facing gluts of vaccine doses.

One common response to this is mandatory vaccination. The California university systems, and 1 in 10 U.S. universities and colleges, almost all in blue states, have announced that vaccination would be required for anyone attending in the fall. Even where demand for vaccination exceeds supply, vaccinations are being made mandatory. In March, the government of Galicia in Spain required vaccinations for adults, subjecting violators to substantial fines. Italy has mandated vaccinations for care workers.

Mandating vaccination may be unavoidable in some cases. But our new evidence from Germany suggests that it could hurt voluntary compliance, prolonging the pandemic and raising its social costs.

Survey evidence from Germany shows growing resistance to enforced vaccination

We looked at the attitudes of a representative sample of 2,653 Germans surveyed during the first and second German lockdowns, in April-May and in October-November of 2020. Respondents were asked: “If there is an approved vaccine against the coronavirus: To what extent would you agree to be inoculated yourself if: … vaccination is strongly recommended by the government but remains voluntary? … vaccination is made mandatory and checked by the government?”

More than two-thirds of respondents supported (either fully or somewhat) voluntary vaccination in both waves. But the fraction fully supporting vaccinations if they were enforced dropped from 44 percent to 28 percent over the same period, even though the infection rate had increased 15-fold between the two waves of our survey.

Trust in government and in science is critical for vaccination acceptance

The most important factor predicting vaccine willingness (and how it changed between the two waves) was whether the respondent trusted public institutions. We measure this by averaging respondents’ expressed trust in the federal government and specifically in its information about coronavirus, as well as trust in state governments, science and the media.

We found that the fall in support for enforced vaccination was disproportionately high among those whose public trust had also declined. The effect is substantial, accounting for a large fraction of the observed reduction in support for enforced vaccinations.

But the growing opposition to enforced vaccination was not a result of increased skepticism about the vaccine itself, or a general reduction in trust, neither of which changed on average between the two waves. About as many people became more trusting as lost trust. Our survey evidence, instead, points to increased opposition to enforcement itself because it “restricts your freedom.”

Mandated vaccinations may provoke distrust

There are three reasons that we think may explain why a government mandate may make people more resistant to vaccinations. First, people resist being manipulated. Behavioral experiments have shown that positive incentives (carrots) and controlling or negative incentives (sticks) often backfire — producing the opposite of their intended result — because they provoke what economists term “control aversion.”

Second, people may be affected by what psychologists call “moral disengagement.” This happens when people think that the decision is predetermined by explicit rules or incentives, so that their own ethical convictions are not relevant. This is one of the reasons traffic engineers now consider traffic lights to be more hazardous than roundabouts. The political scientist Elinor Ostrom described how good citizenship can get crowded out when the government takes responsibility for doing the right thing.

Something like this may be happening with vaccination, with enforced vaccination crowding out altruistic motivations. In the second wave of our survey, those who were more altruistic (they said they were “willing to help others”) were also more likely to support voluntary but not enforced vaccinations.

Finally, enforcement may crowd out intrinsic motivation by diminishing trust. Citizens may take enforced vaccination as evidence that the government thinks people would not get the vaccine if they have a free choice. This may suggest either that there is something wrong with the vaccine, or that the state does not trust its citizens to respect a social norm of protecting others by getting vaccinated. The citizen’s tit-for-tat response to being distrusted may be to oppose the government’s vaccination program.

Effective vaccination policies can exploit herd behavior to achieve herd immunity

Resistance to mandated vaccinations could lead to a downward spiral. Initial public distrust could limit the numbers willing to be vaccinated, prompting mandatory vaccination, kindling further distrust of government and science. Alternatively, if the government were to start off by mandating vaccinations, our results suggest that this might squander a substantial initial willingness to be vaccinated if it were voluntary, forcing a more expansive vaccine mandate and further heightening public distrust.

In other words, implementing a policy may change citizens’ beliefs and preferences in ways that undermine the policy. The officials in charge of vaccines face a specific version of this problem: Mandates may reduce vaccine willingness.

And voluntary policies may be sufficient even if initial vaccine willingness is limited. In our recent PNAS article, we model how herd behavior can propel a population from vaccination hesitancy to a herd immunity target.

The reason is that those initially volunteering to be vaccinated send a positive signal about their willingness to cooperate and their belief in the vaccine’s safety and effectiveness, which may influence others. By contrast, complying with enforced vaccination sends a much weaker signal, dampening the positive herd effects of others having been vaccinated.

To be clear: Our research does not show that policymakers should avoid vaccine mandates. Legally required vaccination against measles and other diseases play an essential role in public health policies around the world. Instead, our findings concern the design of policies — showing how mandated policies may provoke pushback, how enhancing public trust might mitigate this, and how, under a voluntary policy, herd behavior may be sufficient to reach herd immunity.

Katrin Schmelz is a behavioral economist and psychologist at the Cluster of Excellence on “The Politics of Inequality” at the University of Konstanz and the Thurgau Institute of Economics.

Samuel Bowles directs the behavioral sciences program at the Santa Fe Institute and is the author of “The Moral Economy: Why Good Incentives are No Substitute for Good Citizens” (2016).

Updated July 12, 2021 to clarify author affiliation.