A growing literature documents how religious institutions shape behavior through social influence, but less is known about what happens when religious movements gain political power and use the tools of government to advance their agenda. We use a regression discontinuity design on close mayoral elections in Brazil to show that mayors from parties institutionally tied to Pentecostal denominations increase teenage fertility 3 per 1,000 higher (a 40% increase). This effect appears for cohorts exposed to middle school during the administration. Consistent with a school-based mechanism, we find that the likelihood that municipal schools offer sexual education programs falls by 12.5 percentage points, with no changes in state schools outside mayoral control. We also find elevated STD rates, and higher middle school dropout rates, while slightly older cohorts show no effects. Results are not explained by changes in contraceptive availability in public clinics, pointing to sexual education as the primary mechanism. We also find no effects from other right-wing parties, indicating the importance of institutional links to Pentecostal parties.
Alex: Welcome to another episode of ResearchPod. Sam, what study are we diving into today?
Sam: This is research by Marcela Mello and João Garcia on Pentecostal mayors in Brazil and their effects on teenage pregnancy. The core finding is that when mayors from parties tied to Pentecostal churches win elections, teen birth rates rise about 40 percent higher among girls going through middle school during their term. It's not just social influence from religion—it's how they use government control over local schools.
Alex: So this paper is basically asking what happens when religious groups get real political power over schools, rather than just preaching from the pulpit?
Sam: Yes, exactly. Past studies look at religion shaping behavior through community norms or beliefs—like how church teachings might discourage premarital sex socially. But here, the researchers examine Pentecostal mayors who can directly change school policies, such as cutting sexual education programs in the schools they control.
Alex: Right, and in Brazil, mayors have that kind of hands-on power over local schools?
Sam: That's key to the setting. Brazil's towns elect mayors who appoint school principals and decide on things like health classes, which aren't required nationwide. These Pentecostal parties—PRB and PSC—have deep church ties, so their mayors push conservative views on sex ed, unlike other right-wing groups. The study compares towns where these candidates barely win or lose close elections, isolating the mayor's impact.
Alex: Okay, so the puzzle is pinning down if it's really the schools causing higher teen births, not something else in those towns.
Sam: Precisely. They focus on girls aged nine or ten at election time, who get full middle school exposure under the new mayor, versus older girls already past that stage. Birth rates jump three per thousand—a 40 percent increase—for the younger group, with other signs like higher syphilis and dropouts, but no changes in contraceptives or clinics. This points to schools as the main channel.
Alex: So the younger girls see this birth rate jump of about 40 percent, plus higher syphilis and dropouts, while clinics stay the same. But how do we know the mayors are actually changing what's taught in those middle schools?
Sam: Mayors in these Brazilian towns run the municipal schools, which cover middle school for kids around 10 to 15 years old. They can pick the principals—about 40 percent get the job through political choice, not tests or votes—which lets them install people who share their views on things like health classes. Principals then shape the daily lessons, deciding how much time goes to optional topics like preventing pregnancy. The study finds Pentecostal mayors cause 21.7 percentage points more principal turnover in municipal schools, leading to a 12.5 percentage point drop in sex education classes there.
Alex: Principal turnover as the lever—that's a direct way to shift the curriculum without national rules getting in the way. And state schools don't see that?
Sam: Correct. State schools in the same towns handle high school and pick their own leaders through state processes, so mayors can't touch them. The Prova Brasil survey—questionnaires filled out by principals every couple years—shows no drop in sex ed for those state schools under Pentecostal mayors. This split confirms the change happens only where mayors have control: municipal middle schools serving the younger girls.
Alex: Huh, so the unaffected state schools act like a built-in check. What about the parties themselves—do PRB and PSC stand out in pushing this family-over-schools view?
Sam: PRB grew directly from the Universal Church of the Kingdom of God leaders in 2005, with the party's structure matching the church's setup for pushing views on sex and family into laws. PSC channels candidates from the Assembly of God, Brazil's biggest Pentecostal group, and its rules match their stances, like protecting life from conception. Both prioritize moving moral topics like sex ed from classrooms to parents, as seen in their support for bills and local statements saying schools should stick to math and science.
Alex: That explains the consistent pressure when they win close races. So that path leads to higher births—but are there other signs tying it back to less sex ed, like infections or kids leaving school early?
Sam: Yes, the study looks at syphilis cases—sexually transmitted infections reported to national health tracking. Among the younger girls fully exposed in middle school, rates rise notably under Pentecostal mayors, while older girls see no change. Dropout rates also climb in the final years of municipal middle school for those same girls.
Alex: Huh—syphilis up, dropouts up for the younger group. And clinics? No cuts there to explain it?
Sam: Evaluators checked municipal health clinics for supplies like pills, condoms, and emergency options—no differences under Pentecostal mayors. Fetal deaths and hospital abortion records show nothing notable either, which fits since the main issue is prevention through school classes, not endings. This rules out health services as the driver.
Alex: Right, so schools are the key channel—the rest stays steady.
Sam: Exactly. Dropouts hit municipal schools only, where principals were swapped out, and mostly in 8th and 9th grades when births start appearing. The close-election balance on town traits like income and evangelicals further supports that it's the mayor's school control making the difference.
Alex: That balance in close elections sounds solid, but how do they double-check no one gamed the vote counts right at the edge?
Sam: They start by looking at the spread of election margins around the exact 50 percent cutoff—like counting how many races ended at 49.9 percent versus 50.1 percent. If politicians or voters fudged results to push just over or under, you'd see a pile-up or gap in one spot, but the graph shows a smooth line with no jump. Researchers call this a density test, and it passes easily, meaning wins were as good as random near the line.
Alex: So no manipulation skewing the groups—fair coin flip for those towns. And the towns or candidates themselves match up okay on basics like size or money?
Sam: Yes, graphs compare barely-winning Pentecostal spots to barely-losing ones on things like population size, city-vs-country mix, reading levels, sanitation, income, even share of Pentecostals—all line up closely. Candidates match too on age, gender, marriage rates, with just a small blip on college education that isn't a big deal. This balance backs the idea that the close wins didn't sort richer or poorer places systematically.
Alex: Huh, those checks make the comparison feel reliable. But to really nail schools over other town changes, they test older girls with no middle school overlap, right?
Sam: Exactly, and it works. For girls 16-17 at election—already past school under the mayor—no birth rate shift, even though their baseline is higher; estimates hover near zero across setups. A pre-election check on the younger group four years earlier, under old mayors, also shows no difference. These placebos rule out general town shifts like clinic cuts or norms changing everything.
Alex: So effects only hit the school-timed group. Do they show when exactly the births start climbing?
Sam: They track year-by-year for the 9-10 girls: nothing in years 1-2 at ages 10-12, then up from year 3 at 12-13, peaking at over 6 extra per thousand by year 5 at 14-15. Older placebo cohort shows random wiggles, no pattern. Timing matches max exposure to swapped principals and fewer classes during key teen years.
Alex: That gradual build pins it to school years precisely.
Alex: Okay, that timing fits middle school perfectly. But to confirm it's the sex ed cut itself—do they measure how much those classes actually drop in the mayor-controlled schools?
Sam: They use surveys from school principals every few years, asking what optional classes they offer. In municipal middle schools under Pentecostal mayors, the share drops by 12.5 percentage points—about an 18 percent cut from baseline—while state schools in the same towns show no change at all. This split shows the effect comes from mayors appointing principals who skip those classes, not some town-wide shift.
Alex: Right, state schools as the control right next door. And those new principals—any direct sign they're picked for their views?
Sam: Surveys from 2009, 2013, and 2019 show a 21.7 percentage point jump in municipal principals with under two years experience right after elections—meaning fresh appointments. State schools see none of that. It points to mayors installing aligned leaders who then deprioritize sex ed and similar moral topics.
Alex: Huh, so turnover lets them steer the curriculum selectively. Does that ripple into health outcomes beyond births, like infections?
Sam: Yes, syphilis diagnoses—mandatory reports to health authorities—rise by 0.514 cases per thousand for the young cohort over four years, about three times the low baseline. No change for older girls past school age, and HIV shows nothing. It's rare overall, skewed by a few outbreaks, so most towns see no syphilis uptick, but it fits less prevention in school.
Alex: That selective hit on the school group again. And dropouts—do pregnancies push girls out?
Sam: Dropout rates at middle school end rise 2.8 percentage points in municipal schools for those girls—from a four percent base—none in state schools. It could go either way: pregnancy causing dropouts, or less engagement raising pregnancy risk. Either links back to the curriculum changes under new principals.
Alex: Strengthens the school channel. But what rules out mayors cutting clinic supplies instead?
Sam: Checks on six contraceptives in health centers—pills, condoms, IUDs, emergency options—find no drops, all near universal already. Abortion hospitalizations and fetal deaths show zero change too. Schools stand out as the driver.
Alex: So not health access. One more: is this just any conservative mayor, or the religious tie?
Sam: They test close wins by other right-wing parties—similar conservative scores—and find no birth or sex ed effects, estimates tiny and insignificant. It ties specifically to Pentecostal religious leverage, not generic conservatism.
Alex: So pulling it together, the evidence points to schools as the main pathway, with that 12.5 percentage point drop in sex ed classes under Pentecostal mayors, but only where they control things. Still, the paper notes some limits there—worth unpacking?
Sam: Yes, a key limitation is they can't link individual girls' births directly to which school they attended—records show residence but not enrollment history. So while the patterns align compellingly—younger cohorts exposed during school years, effects only in municipal schools, gradual buildup—it relies on that indirect fit rather than tracking each student. The sex ed measure also captures just whether schools offer any classes at all, not depth or quality of lessons.
Alex: Right, so it's a strong signpost, but not the full picture of every classroom change. And those other school shifts—like new principals bringing informal messaging on family roles or closer church ties—could amplify it?
Sam: Exactly, those are complementary channels the data supports indirectly: turnover installs aligned leaders who might stress traditional norms beyond formal classes, or foster peer expectations around early marriage. Syphilis rising about three times baseline for the exposed group fits less prevention knowledge leading to riskier choices, not access issues. It's all tied to the school environment reshaping behavior.
Alex: Huh, that broader reorientation makes sense—targeted at sex topics while other classes stay steady. But scope-wise, these are smaller towns around 20,000 people—does it carry to bigger cities?
Sam: The findings fit best for those smaller municipalities where mayors have real sway over schools and Pentecostal groups are embedded locally—about the median size, but excluding spots with 60 percent of Brazil's population. Urban areas have tighter rules, more state or private schools, and easier info access, so effects might weaken there. Still, it highlights decentralization's risks: local control lets ideological groups reshape services like sex ed, hitting human capital via pregnancies, dropouts, and infections.
Alex: A meaningful caution on how power at the local level can have these demographic ripples, specific to religious ties over plain conservatism. Thanks, Sam—that's a clear window into the mechanics.
Sam: Precisely, and with dropouts up, it suggests long-term costs like lower earnings worth tracking ahead. This study offers solid quasi-experimental evidence on religious politics steering public services in decentralized setups. Thanks for listening to ResearchPod.