The sheer number of Rohingya Muslims fleeing genocide in Burma—over 10,000 per day since late August—has become too huge to ignore. It’s the reason why U.S. Secretary of State Rex Tillerson will visit Burma on Wednesday. In his briefings on the crisis, Tillerson will likely encounter another question of numbers: the claim, voiced often by Burmese officials and hardline Buddhist monks, that Rohingya Muslim “overpopulation” threatens their country’s Buddhist majority.

“The population growth of Rohingya Muslims is 10 times higher than that of the Rakhine [Buddhists],” said Win Myaing, a spokesperson for Burma’s western Rakhine State, where most of the stateless Rohingya live. That was in 2013, when the state passed a controversial two-child limit law that applied only to Muslims. Just last month, an administrator of a “Muslim-free” village outside Yangon told The New York Times, “[Rohingya] are not welcome here because they are violent and they multiply like crazy, with so many wives and children.” The motto of Burma’s immigration ministry is, “The Earth will not swallow a race to extinction, but another [race] will.”

The specter of an exploding Rohingya population—fueled by incendiary content on social media—has been weaponized in Burma’s ethnic cleansing. It haunts everyone from the military and Buddhist nationalists to ordinary Burmese citizens. I myself often come across this reprinted claim. Despite its clear attribution to sources like extremist monks, I started to wonder if it was true. It’s not.

In fact, according to a study published in 2013, there was a net outflow of Rohingya from Burma after 1950—and that was before the unprecedented exodus of the last three months. Moreover, Burma’s Muslim population has been stable at around 4 percent since the 1980s, according to the country’s own census.

The trope of Muslim overpopulation is reliably powerful anywhere in the world where there is a sizable Muslim immigrant or minority population, from India to Western Europe.

Hindu nationalists often fan anxiety about Muslim population growth; the proportion of Muslims in India grew about 0.8 percent between 2001 and 2011, to 14.2 percent. “If this remains the situation, one should forget about their existence in one’s own country by 2025,” said the leader of a major Hindu nationalist organization last year. But the fertility gap between Muslims and Hindus in India is narrowing fast, and the greatest birthrate disparities are between states, not religions: Hindu women in the very poor state of Bihar have about two more children each than Muslim women in more developed Andhra Pradesh.

Similar concerns echo across countries like France, Germany, the U.K., and the Netherlands. Although Muslims make up less than 10 percent of the total population in each of these countries, perceived overpopulation has been at the center of anti-immigration discourse. About 7.5 percent of France is Muslim, yet on average French people believe Muslims constitute about one in three people in the country. Although Muslim women in Western Europe do currently have more children than their non-Muslim counterparts, research shows that European Muslims’ fertility rate is also declining much faster, so their fertility rates will likely converge over time. (However, in this context, fertility isn’t the only issue; a wave of Muslim immigration over the past few years has reinforced some Europeans’ concerns about Muslim population growth.)

Why does the overpopulation myth persist worldwide, even though it’s typically demonstrably false (like in Burma) or nowhere near the epidemic that its proponents assert (like in Europe and India)? It’s true that the global Muslim population is growing, and fast. But it’s not growing at the same speed across regions. And the trope seems to have the most power not where Muslim populations are actually growing the fastest—like sub-Saharan Africa—but in places where they are culturally distinct minorities.

There’s nothing inherent in Islam to link it to higher fertility—in fact, it’s not a particularly natalist, or pro-birth, religion. Eight of the nine classic schools of Islamic law permit contraception. Many Muslim states, including Pakistan, have supported family planning. The growth of the global Muslim population was, according to a 2011 Pew Center report, due to both a “youth bulge”—an unusually high number of young Muslim people, which peaked around 2000— and a higher overall fertility rate for Muslim women as a group.

On the latter point, a major takeaway of the Pew report (and its companion from this year) is that fertility has much less to do with religion and much more to do with economics, social services, women’s empowerment, and conflict. The fertility rate across all 49 Muslim-majority countries fell from 4.3 children per woman in 1990-95 to about 2.9 in 2010-15. This was still higher than the global fertility rate in 2015, but it’s a strikingly fast drop given the fact that it took some Western European countries nearly a century to transition from six children per woman to three.

The claim about Muslim overpopulation falls apart in fascinating ways when examined more closely. The fastest fertility drop in modern history happened in the Islamic theocracy of Iran. In 1950, Iranian women had about seven children each; today they have about 1.68, fewer than Americans. What changed? In 1989, the country’s leaders realized that the the high birth rate was straining the young republic. In response, the Supreme Leader issued fatwas encouraging birth control and contraception, and the Health Ministry propagated family planning counseling, rural health centers, and contraceptive distribution across the country. Iran also made girls’ education a development priority as it sought to rebuild civil society after the Iran-Iraq War, which ended in 1988, so more girls than ever started to attend (strictly gender-segregated) schools. Everywhere, there is an inverse relationship between years of schooling and fertility rates.

In the world’s largest Muslim-majority country, Indonesia, fertility rates dropped between the 1960s and the 1990s, from about 5.6 children per woman to 2.3, as the Suharto dictatorship instituted a vigorous, centralized family planning program and made improvements to girls’ education. Those government services were decentralized after democracy came to the archipelago in 1998 and, predictably, fertility rates have been creeping up again. Today, Indonesia’s majority-Christian but less developed eastern provinces have a higher birthrate than the more developed, Muslim-majority western ones—a testament to the correlation between economic development and fertility.

But it’s unlikely that these, or any, facts about Muslim demographics will change minds anywhere. In a demographics report, Nicholas Eberstadt and Apoorva Shah, researchers at the American Enterprise Institute, write: “There remains a widely perceived notion—still commonly held within intellectual, academic, and policy circles in the West and elsewhere—that ‘Muslim’ societies are especially resistant to embarking upon the path of demographic and familial change that has transformed population profiles in Europe, North America, and other ‘more developed’ areas.”

In places like Burma, where the Muslim overpopulation trope is now deeply rooted, facts may have even less sway. “Repeated exposure is a big factor in determining the ‘stickiness’ of misinformation,” said Sander van der Linden, a social psychologist at the University of Cambridge. He pointed to the mere-exposure effect, a tendency to develop a preference for things simply because we’re familiar with them, and the illusory-truth effect, a tendency to believe information after repeated exposure. “Both point to the fact that the more a falsehood is repeated, the more likely people are to believe it,” he said, adding that political leaders have long understood this concept on an intuitive level. “Consider the ‘big lie’ law of propaganda: ‘If you tell a lie big enough and keep repeating it, people will eventually come to believe it.’ … We’ve seen many ‘big lies’ fly by in current debates, and some people continue to believe them.” (After reading a few hundred stories on the Rohingya crisis, I was almost one of them.)

What’s more, in Burma, where open internet access is less than a decade old, news content does not necessarily get evaluated critically. As one observer put it recently, for the Burmese, “The entire internet is Facebook and Facebook is the internet,” so there is plenty of opportunity for repeated exposure to incendiary fake news about the Rohingya.

The consequences of the Muslim overpopulation myth are chilling. They’re also sadly ironic, because the myth has likely been counterproductive for its propagators. Given the socioeconomic underpinnings of fertility, the targeted persecution in Burma may have made high Rohingya birth rates a self-fulfilling prophecy. There are only a few factors that reliably decrease the fertility rate in any developing country: more schooling for girls, the expectation that one’s existing children will survive (due to healthcare and freedom from conflict), access to contraception, and job opportunities for women. By denying Rohingya women all of the above, the Burmese military may be creating the exact result that its supporters feared in the first place.