Vaccination incentives can backfire

Of all economic claims, the idea that “people are more likely to do what you ask if you pay them” is probably one of the least controversial. After all, paying money in exchange for something is one of the most basic features of commercial activity. Nonetheless, some policymakers were cautious about using financial incentives to encourage vaccination during the COVID-19 pandemic.  Many wondered if paying people to get vaccinated was an atypical case and could be riskier than more traditional approaches, such as information campaigns and vaccination mandates.

A recent spate of formal evaluations has shown that most incentives either increased vaccination rates during the pandemic or, if they didn’t work, at least had no harmful effects. For instance, results from a paper recently published in the journal Nature was entitled “Financial incentives for vaccination do not have negative unintended consequences,” based on a randomized controlled trial in Sweden. This and other studies from the COVID-19 pandemic add to what was already a large literature on the positive effects of vaccination incentives. For example, in 2010, Banerjee and co-authors found that financial incentives significantly increased vaccination rates among the poor in India. A rigorously evaluated vaccination incentive program in Nigeria is one of the top-rated charity recommendations from GiveWell, an organization that assesses the value of aid programs.

So, in 2021, as COVID-19 vaccine uptake slowed in Central Asia, discussions among government experts and development partners quickly turned to the question of using financial incentives.  However, since large-scale programs had not been used in the region before, the World Bank team chose to test the idea before endorsing it. We conducted a quick evaluation using three nationally representative household panel surveys that were already being collected as part of a project called Listening to Central Asia.

Our approach was simple. We conducted a randomized controlled trial in Uzbekistan, Tajikistan, and Kazakhstan, testing the impact of different levels of hypothetical financial incentives on individuals' stated intention to get vaccinated compared to a control group that was offered no financial incentive. The value of the incentive ranged from the local currency equivalent of about $5 up to $50. Beginning in May 2021, a total of 6,783 individual responses were collected—2714 in Uzbekistan, 2267 in Tajikistan, and 1802 in Kazakhstan.

Our study found that offering financial incentives actually reduced vaccination intentions by about 19% compared to the control group, with the strongest negative effect observed in Uzbekistan and Kazakhstan. Country-level results ranged from no meaningful effect on vaccination intentions (Tajikistan) to a decline of up to 22 percent (Uzbekistan and Kazakhstan).

We also asked respondents about their views on vaccination incentives, and found that the majority disapproved of them, particularly in the countries where the experiment showed negative effects. About three quarters of respondents in Uzbekistan actively opposed providing cash incentives for vaccination. In Kazakhstan, 58 percent said they disagreed with paying incentives. Only in Tajikistan, where the experiment failed to find any systematic effect on vaccination intentions, did just 18 percent say they disagreed with paying incentives.

The results were a stark contrast with the well-established positive effects of monetary incentives studied elsewhere and suggest that the effectiveness of vaccination incentives is not universal. Several similar studies—including from Rutschman and Wiemken, and Robertson et. al.—have suggested heterogenous effects depending on norms and customs, which can vary both between countries and within countries with diverse communities. In novel contexts where they have not been used before, incentive payments could also signal inferiority or disutility. For instance, a person might ask “nobody has ever offered me money for a routine vaccination before, what is so bad about this vaccine that they have to pay me to take it?” Following our study, several public health experts in Central Asia offered variations of this explanation.

Our study suggests that behavioral responses to financial incentives can be complex and context dependent. What works in one setting may not work in another.  Also, cultural norms and beliefs can play an important role in shaping individuals' responses to incentives. As such, incentive programs should be carefully designed and tailored to specific contexts to maximize their effectiveness. The risk of a “backfire” effect—such as that we found in Kazakhstan and Uzbekistan—implies that it is prudent to pre-test incentive programs when they will be used in novel settings.