The Washington PostDemocracy Dies in Darkness

It ended in 1767, yet this experiment is still linked to higher incomes and education levels today

The ruins of the Jesuit mission at Santisima Trinidad del Parana in Paraguay. Shown here in 2010, the mission was built in 1706. (NORBERTO DUARTE/AFP/Getty Images)

Investments in roads and cities can shape an economy for hundreds of years. Now economists are showing how one-time investments in education can endure every bit as long.

How 2,000-year-old roads predict modern-day prosperity

Jesuits arrived late to the Guarani people’s homeland where Brazil, Paraguay and Argentina meet. But their missions thrived until 1767, when King Charles III of Spain expelled all Jesuits from the Spanish Empire.

Yet even 250 years later, people living near the ruins of Jesuit missions complete 10 to 15 percent more years of education and earn 10 percent more than residents of equivalent towns without missions, according to a study published in the Quarterly Journal of Economics.

Measuring an intangible asset across centuries requires clever techniques. It also requires the perfect natural experiment. University of British Columbia economist Felipe Valencia Caicedo relied on both.

Valencia’s analysis is among the most striking of a surge of studies that show how returns from education and vocational training span generations and even centuries.

A historical experiment

You would need a randomized experiment to truly disentangle education from the confounding factors that usually come with it. You would scoop up entire communities and dump them in an isolated region. You would divide them into equally matched treatment and control groups, and observe everyone carefully for an epoch or two.

Valencia couldn’t do that, for obvious ethical and practical reasons. But he found the next best thing. The 34-year-old Colombian, who studied economics at Brown, Yale, the London School of Economics and Pompeu Fabra University in Barcelona, has long been fascinated by Latin America’s colonial past. He zeroed in on this isolated archipelago of Jesuit missions because they were free of complications that would make it impossible to tease out the effect of missions founded in, say, the megalopolis of Sao Paulo.

Founded in 1534, three decades before it built its first South American mission, the Jesuit movement, with its technology and methodology, was advanced for its time. Importantly for this study, its members focused on education. Their Guarani missions were seen as an egalitarian, utopian experiment, earning such famous fans as Voltaire and Rousseau.

At the 30 missions Valencia studied, the Jesuits taught indigenous men and women of the Guarani tribe blacksmithing, arithmetic and embroidery. Those efforts stopped after the Jesuits were expelled.

The clean break made the perfect starting point for an economic analysis.

“Moral considerations aside, for us in economics colonialism is a nice historical experiment to study things that we care about,” Valencia said. “We are not colonialists … but we can study these effects.”

Measuring the effects

By piecing together data from archives in the Vatican, Spain, Paraguay and Argentina, Valencia found ample evidence that the Jesuits’ training continued to influence the education and society of the Guarani people who lived nearest the missions.

For every 100 kilometers closer to a mission, poverty rates fall 10 percent and schooling rises by about 0.7 years. The effect was sharper 75 or 100 years ago, historical data shows, but as national literacy approaches the maximum of 100 percent, differences flatten out.

For variable after variable, Valencia found the same positive trend. That isn’t an endorsement of the Jesuits’ methods; he said his work shows the enduring power of investing in human beings, not the enduring power of colonialism or Catholicism.

“This paper is not saying missions are good,” Valencia said. “This paper is not saying Catholicism is good or didn’t have any negative effects. This paper is saying even though people were converting indigenous people to Catholicism, while they were doing that they were also teaching them skills.”

Princeton economist Leonard Wantchekon, working with Marko Klašnja of Georgetown and Natalija Novta of the International Monetary Fund, used different methodology but reached a similar conclusion in an analysis of the long-term effects of early 20th-century missionary schools in Benin.

Their analysis, published in the same journal, found the benefits of schooling to be contagious — one possible reason the benefits of education persist for generations.

“Children whose parents did not attend school, but were living near those who went to school, did as well as those whose parents did attend school,” Wantchekon said. “In other words, aspirations played an important role in intergenerational mobility."

University of Michigan economist Hoyt Bleakley has published several influential works that measure the long-term fallout of circumstances ranging from hookworm eradication in the American South to river portages on the East Coast. He said Wantchekon’s and Valencia’s work showed the durability of some human capital investments.

“These fairly early human-capital-building institutions seem to leave a pretty significant footprint for generations,” Bleakley said. “Maybe for centuries.” According to Bleakley, it’s difficult to separate the effect of education from that of the cultures and institutions that tend to be adopted by educated people.

Isolating causes

Any armchair statistician can tell you that these findings are nigh on useless if they reflect a third variable, such as desirable location or high population density, that explains both mission location and education level. Which is why Valencia — like all economists who do this type of research — went to extraordinary lengths to tease apart correlation and causation.

Because they arrived after the Franciscans and others claimed prime mission territories, the Jesuits were forced to build their missions in unremarkable places. Today, the settlements near the missions are no more densely populated than the areas around them. Valencia’s analysis of satellite data also shows they don’t have temperatures, terrain or climate more conducive to growing crops. They’re not closer to rivers or the coast, either.

Valencia found a number of natural control groups. People who immigrated to these communities, for example, don’t show the same education trends as the native-born.

The Franciscans helped, too. The mendicant order had been in the area longer than the Jesuits. It wasn’t expelled. But the Franciscans focused on poverty and health, rather than education and vocation. Valencia found that, perhaps as a result, families’ economic and educational outcomes didn’t vary based on their distance to that order’s missions.

Maybe the Jesuits had a preternatural ability to select locations that would later attract educated people? We can rule that out, too. There were numerous locations where the Jesuits started a mission but quickly abandoned it for reasons unrelated to its economic viability. Even those locations, compared with similar ones with long-lasting Jesuit missions, produced no measurable effect.

Finally, Valencia ruled out institutions, which development economists consider to be a likely driver in cases like this. Brazil, Argentina and Paraguay had wildly different governments, data sets and institutional experiences over the centuries he studied, yet they all showed similar trends.

Pinpointing mechanisms

Valencia didn’t just rule out other possible explanations. He also established a few reasons that areas near the missions have tended to succeed at higher rates, even today.

People nearer the missions were more likely to pass knowledge from generation to generation. Valencia’s survey showed this not just for skills taught by the missionaries, such as embroidery, journal-keeping and accounting, but also for traditional medicine and folklore not directly related to the Jesuit mission.

Historical descriptions of the missions emphasize vocational training that went beyond agriculture to trades and service industries. Today, the areas near the missions are more industrialized than their neighbors, and they tend to have more people employed in industries related to the professions emphasized by the missionaries, such as blacksmithing and teaching.

This shouldn’t imply that the differences remained constant, however, or that history is destiny. Areas near the missions embraced innovations more readily than their peers. Between 1996 and 2006, farmers nearer the missions in Brazil were faster to plant new, genetically engineered soybeans than those farther away.

“It’s not that things are immutable,” Valencia said. “It’s that precisely because of this historical human capital shock, that people can change faster and can adapt faster to change.”

The enduring educational and cultural imprint of that historical shock — an educational mission cut short about a decade before the U.S. war for independence — is strong evidence that investing in people produces long-term returns.


An earlier version of this story contained an incorrectly captioned image of the Jesuit mission of Santisima Trinidad Del Parana after a wire service error. This has now been corrected.