Infectious disease epidemics disrupt communities temporarily but may also lead to long-term changes in children’s human capital accumulation (Almond, 2006, Beach et al., 2022a, Meyers and Thomasson, 2021). Measles is the world’s leading childhood cause of vaccine-preventable illness and was until the 1960s one of the most common sources of epidemics in the United States. Before the measles vaccine, measles infected 80 to 95 percent of American children by age ten and resulted in 48,000 hospitalizations, 1,600 cases of severe brain damage, and 400 child deaths annually (Centers for Disease Control and Prevention, 2020, National Communicable Disease Center, 1966b, National Communicable Disease Center, 1966a, National Communicable Disease Center, 1967a). These most severe health costs of measles are well documented, but less evidence exists about the long-term effects of measles control for educational outcomes.

This paper shows that the first mass vaccination campaign against measles in 1967–1968 resulted in a modest but statistically significant increase in the educational attainment of men in the affected cohorts. To identify these effects, we take advantage of the Measles Eradication Campaign launched in 1967 that reduced reported measles cases by 90 percent within two years. The immunization campaign intended to increase the public provision of the measles vaccine and to achieve universal measles immunization coverage (Sencer et al., 1967). The vaccine was more expensive than immunizations available for other diseases (Colgrove, 2007), and surveys conducted before the campaign revealed that vaccination rates were mainly increasing in wealthy suburban neighbourhoods (National Communicable Disease Center, 1966b, Cliff et al., 1993, Pendergrast, 2010). Consequently, the mass vaccination campaign became part of President Johnson’s War on Poverty and the first nationwide federally subsidized immunization campaign, after the federal government created formal immunization assistance in 1962 (Johnson et al., 2000, Hinman et al., 2011).

Publicly provided preventive health inputs may affect human capital accumulation in several ways (Bärnighausen et al., 2014, Lleras-Muney and Cutler, 2014, Almond et al., 2018, Dupas, 2014, Baird et al., 2016). Among them, at least two mechanisms suggest a potential causal link between mass vaccination against measles and educational attainment. First, due to immunization, some children never contracted measles or its complications. They also did not suffer from measles-induced immune system suppression (Mina et al., 2015, Mina et al., 2019) and may have, therefore, been less ill during their childhood. Healthier children, in turn, may have been less absent from school or learned better while at school (Miguel and Kremer, 2004). Second, after mass vaccination reduced measles outbreaks, primary school children regardless of immunization or disease status were less exposed to measles control efforts, such as quarantines.

To identify the educational impacts of mass vaccination against measles, we use a difference-in-differences strategy that employs variation by cohort and geographic area. First, we take advantage of the sudden reduction in measles cases due to the Measles Eradication Campaign of 1967–68. The reduced measles exposure was more likely to benefit younger cohorts because they were less likely to have contracted measles by the time of the immunization campaign. Measles can only be acquired once, and either a measles infection or immunization after age one confers immunity. According to survey evidence, vaccination rates increased for the cohorts targeted by the campaign in line with the regular age profile of measles infections. Thus, different proportions in each cohort of children were vaccinated or infected with measles.

Second, we exploit variation in treatment intensity by geographic location. We compare the evolution of educational attainment for children born in states that experienced higher reported measles incidence right before the campaign to those cohorts born in states that experienced lower measles incidence. Reported measles incidence declined more in the high baseline measles incidence states. Therefore, measles exposure changed more for cohorts of children born in the states that experienced a higher measles burden prior to the campaign. Thus, younger cohorts were likely to benefit more from the campaign in the states with high measles exposure, compared to older cohorts.

Before the cohorts affected by the campaign, the average educational attainment was lower in the states with a higher measles burden but developing similarly to the states with less measles exposure. Subsequently, the average educational attainment improved for the cohorts affected by the campaign and, notably, the gap in educational attainment narrowed between high and low measles exposure areas.

Our main estimates suggest that mass vaccination against measles increased educational attainment. Moving from the 5th percentile (baseline measles incidence 45 reported cases per 100,000) to the 95th percentile of our treatment intensity variable (baseline measles incidence 2,327 reported cases per 100,000), years of education increased on average by 0.08 for the cohorts targeted by the vaccination campaign or born after the large decline in measles incidence. Our estimate can be interpreted as the effect of reducing measles incidence from a state with a high measles burden right before the campaign to close to zero, which is in line with the rapid reduction of measles incidence for these cohorts. We further show that these results are driven by the educational attainment of men, who gained on average 0.12 years of education. For women, our point estimates are smaller and most of them are not statistically significant. Our results are robust to a different definition of our treatment intensity measure, to controlling for several baseline characteristics of states, and to a placebo analysis with a different disease. In addition, we find tentative evidence that the college graduation rate of men increased by approximately two percentage points on average.

Our paper is the first to identify impacts of reduced childhood measles exposure on educational attainment in adulthood. The most closely connected literature explores the educational impacts of immunization policies (for tuberculosis Bütikofer and Salvanes (2020), for polio Serratos-Sotelo et al. (2019), for tetanus Canning et al. (2011), for vaccines against multiple diseases Alsan (2017) and Lee Luca (2016)). Compared to this research, the contribution of our analysis is that we are able to isolate the effect of a mass vaccination campaign against one of the most common childhood diseases on completed education in adulthood. Measles continues to pose a major health risk to millions of children in many low-income countries but has also resurged in places that had previously eliminated endemic infection, such as the United States and many European countries (World Health Organization, 2019).1 Compared to prior research on measles and educational outcomes (Anekwe et al., 2015, Nandi et al., 2019), we are able to take advantage of a policy-driven reduction in measles exposure, a large representative sample on a national level, and follow-up until adulthood. In terms of institutional context, the paper closest to our analysis is Atwood (2022), who examines the introduction of the measles vaccine in the United States and finds that it increased earnings and employment in adulthood for the cohorts born after the licensing of the vaccine. Instead, we focus specifically on the Measles Eradication Campaign, a federal policy initiative to achieve widespread coverage with the vaccine. Reported measles cases first reached unprecedentedly low levels in 1967, the first year of the campaign. Prior to this, the vaccine had failed to reach most poor families. Thus, our analysis is consistent with the hypothesis that the public provision of the measles vaccine was a significant element in reaching the educational benefits we estimate.

Our analysis also contributes to the literature that uses historical disease eradication campaigns and medical innovations as quasi-experimental variation to estimate the causal impacts of reduced disease exposure on human capital (Acemoglu and Johnson, 2007, Bleakley, 2007, Bleakley, 2010, Cutler et al., 2010, Bhalotra and Venkataramani, 2015, Beach et al., 2016, Lazuka, 2020, Battaglia and Kisat, 2021). Our empirical set-up takes advantage of a federal policy response to a medical advance that leads to a sudden and sizeable reduction in exposure to measles soon after the licensing of the measles vaccine. A strength of our analysis is that we are able to employ a direct measure of reported incidence rather than mortality rates used as a proxy for disease incidence in most previous studies. Furthermore, the age profile specific to measles allows us to specify the potential beneficiaries of the eradication campaign in a precise manner.

Additionally, we add to the literature on the long-term economic effects of infectious disease (Almond, 2006, Parman, 2015, Beach et al., 2022a, Meyers and Thomasson, 2021, Beach et al., 2022b, Ager et al., 2022) and childhood health shocks.2 Most of this literature has focused on in utero or infant exposure. In contrast, we contribute to the scarce literature on health shocks of school-age children. Our analysis highlights that disease exposure in the first grades of school may also affect longer-term educational attainment. Finally, we also connect to the literature on public policy and preventive health inputs (Kremer and Glennerster, 2011, Dupas, 2014, Baird et al., 2016, Banerjee et al., 2021).