People are afraid of HIV. Moreover, people around the world are convinced that the virus is easier to get than it actually is. A common misconception is that a single act of unprotected sex with an HIV-positive person guarantees infection. The truth is that it’s not nearly that easy to get HIV—medical research estimates that the transmission rate is actually about 0.1% per sex act (Hollingsworth et al. 2008), and the Government of Malawi says the risk is 5-10% per year (Malawi National AIDS Commission 2009).
Fatalism as a response to high HIV prevalence
One interpretation of these overestimates of risk is that HIV education has been effective. By 2004, fewer than 1% of Malawians had never heard of HIV (NSO and ICF Macro 2010), and many remain concerned about contracting it through unprotected sex. According to the classic risk compensation model, these fears should lead to reductions in unprotected sex (Peltzman 1975).
Unfortunately, the risk compensation story does not seem to be reflected in actual behaviour—at least not in sub-Saharan Africa, where the HIV epidemic is at its worst. Rises in the prevalence of HIV barely affect sexual behaviour in Africa (Oster 2012). In Malawi, where over one in ten people have HIV—the ninth-highest rate in the world—condoms are only used in 12% of all sex acts. If people are so fearful of HIV, why do they not appear to be adjusting their behaviour to avoid the risk of infection? One possible explanation is fatalism: some individuals fear HIV so much that they adopt a fatalistic attitude, responding to higher perceived risks by engaging in more sexual activity, rather than less. This offsets the decreases in sexual activity by the rest of the population, leading to a small average response.
Fatalistic responses can be rational as long as the perceived risk is high enough (Kremer 1996, O’Donoghue and Rabin 2001). The reasoning is as follows: if the risk of each sex act is high enough, and you have at least one exposure, then your chance of already having HIV may be close to 100%. If the perceived probability of contracting HIV regardless of behaviour becomes high enough, the marginal cost of having more sex approaches zero. This means that if individuals believe the risk of contracting HIV is high enough, the effect of risk beliefs on sexual activity will actually switch signs from negative (classic risk compensation) to positive (fatalism). The same logic applies to other diseases, as well; for example, evidence suggests that Americans were fatalistic about COVID-19 during the early pandemic in the US (Akesson et al. 2022).
Providing information on actual HIV transmission rates reduced risk perceptions
I study fatalism in response to HIV using a field experiment in Southern Malawi (Kerwin 2025). Using a village-clustered probability sample of sexually active adults in a single sub-district, I randomly assigned half of the villages to receive information about HIV transmission rates. Providing accurate information on HIV transmission risk marks a significant departure from conventional HIV messaging, which typically emphasises that risk is very high. However, this approach is consistent with the first basic ethical principle of the Belmont Report, which emphasises respecting people’s autonomy (Office of the Secretary 1979).
The Report states that individuals should not be denied the information necessary to make considered judgments unless there is a compelling reason to do so. In the case of information on HIV risk, the evidence in favour of keeping the truth from individuals is limited. Quantitative evidence shows that average responses to HIV risks are very small in Africa (Oster 2012), and qualitative evidence suggests fatalism may be a big problem in Southern Malawi (Kaler 2003). My research is the first to assess what would happen if individuals were provided information on actual transmission rates. The research was overseen and approved by both COMREC, which is the IRB of the University of Malawi College of Medicine, and by IRB-HSBS at the University of Michigan.
The information treatment significantly decreased individuals’ perceived risk of contracting HIV from unprotected sex. It also led to a small average increase in how much sex people have, by about 10%, or around one sex act every six weeks. According to the fatalism model, however, the average effect is misleading; instead, the response will vary depending on individuals’ initial beliefs about HIV risks. To test this implication, I break down my results by people’s baseline risk beliefs. Specifically, I run local polynomial regressions of the outcome on baseline risk beliefs for both the treatment and control groups, computing the difference between the two, and construct confidence intervals using a village-clustered bootstrap.
Figure 1: Treatment effects on sexual activity by baseline risk beliefs
Treatment effects on sexual activity by baseline risk beliefs
Those with very high baseline risk beliefs became less likely to engage in risky sexual activity
Figure 1 shows that the classic risk compensation story applies to most people—the impacts are positive for most of the sample. But, for those whose initial risk beliefs were above roughly 0.8, the treatment effect is negative, implying that lower risk beliefs lead to less risk-taking. These effects are quite large: for the top decile of baseline risk beliefs, risk-taking declines by 49%. Moreover, there are downstream effects on other outcomes: individuals in this high-risk group increase their self-reported HIV testing rate by 24 percentage points, relative to the control group mean of 8%. This is consistent with individuals who previously believed they were HIV-positive reconsidering that assumption and seeking accurate information on their HIV status.
These findings have important implications for the study of risk compensation more generally. We typically study behavioural responses to changes in actual or perceived risks by assuming they are monotonic functions of the change in risk. If that is true, then simple averages provide a useful summary of risk responses. I show that there are important cases where it does not, highlighting the need to examine heterogeneity in risk responses.
These findings also line up with recent research on how individuals respond to information about their HIV status. Ciancio et al. (2025) studies the long-run effects of encouraging people to learn about their HIV status in Malawi during a period where treatment was unavailable. They find that, among people who were HIV-positive, learning one’s status substantially increased death rates beyond the mortality effects of the disease itself. A likely channel for these results is riskier health behaviours. This finding is entirely consistent with fatalism; my research shows that individuals may exhibit this behaviour without actually being HIV-positive due to high perceived transmission rates.
Implications for health policy in a post-PEPFAR world
This paper has important policy implications for both HIV in particular and disease control in general. Until recently, the success of the President's Emergency Plan for AIDS Relief (PEPFAR) seemed to have beaten HIV, turning it into a manageable chronic disease in Africa—one of the greatest triumphs in public health history. Recent cuts to PEPFAR and USAID have put that victory into question. Encouraging safer sexual behaviour is once again important, and, in doing so, we must avoid the temptation to overstate the risks. Looking beyond HIV, ‘scared straight’ policies that overplay risks are likely to backfire for many diseases; public health authorities should thus avoid overselling disease transmission risks.
This article is an updated version of a blog post that originally appeared on the World Bank’s Development Impact Bloghere. It is reproduced with permission from the Development Impact team.
References
Akesson, J, S Ashworth-Hayes, R Hahn, R Metcalfe, and I Rasooly (2022), “Fatalism, beliefs, and behaviors during the COVID-19 pandemic,” Journal of Risk and Uncertainty, 64(2): 147–190.
Ciancio, A, F Kämpfen, H-P Kohler, and R Thornton (2025), “Surviving bad news: Health information without treatment options,” American Economic Review: Insights, 7(1): 1–18.
Hollingsworth, T D, R M Anderson, and C Fraser (2008), “HIV-1 transmission, by stage of infection,” Journal of Infectious Diseases, 198(5): 687–693.
Kerwin, J T (2025), “Scared straight or scared to death? Fatalism in response to disease risks,” The Economic Journal, forthcoming.
Kremer, M (1996), “Integrating behavioral choice into epidemiological models of AIDS,” Quarterly Journal of Economics, 111(2): 549–573.
Malawi National AIDS Commission (2009), "National HIV/AIDS Prevention Strategy: 2009 to 2013."
NSO (National Statistical Office) and ICF Macro (2010), "Malawi Demographic and Health Survey 2010."
O’Donoghue, T, and M Rabin (2001), “Risky behavior among youths: Some issues from behavioral economics,” in J Gruber (ed.), Risky behavior among youths: An economic analysis, University of Chicago Press, 29–68.
Office of the Secretary (1979), "The Belmont Report: Ethical principles and guidelines for the protection of human subjects of research."
Oster, E (2012), “HIV and sexual behavior change: Why not Africa?” Journal of Health Economics, 31(1): 35–49.
Peltzman, S (1975), “The effects of automobile safety regulation,” Journal of Political Economy, 83(4): 677–725.