We’ve long known that, in general, the parts of the country most obsessed by family values are also the most beset by family breakdown. Crimson-red Southern states like Mississippi, Oklahoma and Texas have the highest rates of divorce, while the liberal, decadent Northeast has the lowest. People have tried to explain this phenomenon in a number of ways. Some conservatives argue that the Northeast has low divorce rates only because it has low marriage rates, too. Others suggest that it’s all about class—in addition to being conservative, the South is poor, and poverty is linked to family dysfunction. “It’s a puzzling paradox,” says Jennifer Glass, a sociologist at the University of Texas. “These are places where you would expect the reverence for marriage and social disapproval for divorce to keep couples together.”

Working with Philip Levchak of the University of Iowa, Glass set out to investigate this paradox. Examining America county by county, they found that, even controlling for income, education and rates of nonmarital cohabitation, the link between conservative Protestantism and divorce remains. It looks as if right-wing Christianity itself undermines modern marriage.

“Conservative religious beliefs and the social institutions they create, on balance, decrease marital stability through the promotion of practices that increase divorce risk in the contemporary United States,” Glass and Levchak write in a new paper, “Red States, Blue States, and Divorce: Understanding the Impact of Conservative Protestantism on Regional Variation in Divorce Rates,” which will appear in the next issue of the American Journal of Sociology. Ironically, the very practices meant to shore up marital security in conservative communities end up sabotaging it. By promoting abstinence until marriage, these communities encourage people to marry young. Poor sex education and limited access to contraception for teenagers lead to unintended pregnancies and shotgun weddings. Gender-role traditionalism leads to single-earner families with precarious finances.

Further, Glass and Levchak write, “the effects of personal and community-level conservative Protestant affiliation are additive, meaning that conservative Protestants in strongly conservative Protestant counties have higher divorce risks than conservative Protestants in mainline dominant counties.”

These findings are surprising, since other social science research suggests that couples who attend church together are more likely to stay married. Glass doesn’t dispute this: “Religious belief in general is a good thing for married couples as a shared activity,” she says. But her work suggests that the positives of shared faith fail to outweigh the negatives of fundamentalist culture.

According to their paper, it’s not just believers who are affected—simply living in an area with lots of right-wing evangelicals makes divorce more likely, because the prevailing community norms and institutions affect everyone. The more powerful Christian conservatives get, the worse the problem becomes. “One plausible interpretation of the results is that as conservative Protestant presence increases, elite conservative Protestant influence grows stronger, which results in policies and programs that do little to reduce divorce, but only increase early marriage,” write Glass and Levchak.

“One of the things that happens is that early marriage and parenthood in particular are bad times for very young women to be entering the labor force,” says Glass. “They withdraw from the labor force and withdraw from schooling to take care of their kids.” Meanwhile, she says, “it’s become very, very difficult for young men to support an entire family. Families that are formed early have a really difficult time making ends meet with the human resources they have at their disposal.”

At a time when Republicans are promoting marriage as a panacea for poverty, these findings have important policy implications. Recently, in a speech given to mark the fiftieth anniversary of the War on Poverty, Florida Senator Marco Rubio said, “The truth is, the greatest tool to lift children and families from poverty is one that decreases the probability of child poverty by 82 percent. But it isn’t a government program. It’s called marriage.” Kathleen Parker published a Washington Post column headlined “To Defeat Poverty, Look to Marriage.” Ari Fleischer, George W. Bush’s former press secretary, weighed in with “How to Fight Income Inequality: Get Married,” in The Wall Street Journal.

Now, marriage can be great—that’s why liberals spend so much time fighting for marriage equality. But encouraging people to get married before they’re ready and encouraging them to put off having sex until they wed is a recipe for family instability. “Clearly you can’t put people with few relationship skills and few resources together at a really young age and saddle them with children and expect them to survive,” says Glass.

The blue state model—marriage is delayed; responsible premarital sex is approved—simply works better. That means emphasizing sex education and access to contraception and abortion while letting go of the fantasy of the male-breadwinner family. It means accepting that abstinence until marriage wouldn’t be a useful goal even it was realistic. It means realizing, once and for all, that conservative family values don’t work to conserve actual families.