In the early 1960s, Catholic bishops from around the world met to update Church doctrine for a new era. Reforms made by the Second Ecumenical Council of the Vatican, also known as Vatican II, were meant to foster more inclusive congregations. The most famous allowed priests to celebrate Mass in languages other than Latin.
Academics and certain Church insiders have long contended that Vatican II backfired and instead triggered a worldwide decline in Mass attendance. Now an economics working paper has bolstered that claim with new levels of statistical detail.
“Between 1965 and 2010, we find a striking worldwide reduction in Catholic participation in formal services,” said co-author Robert J. Barro, Paul M. Warburg Professor of Economics. “It cumulates to something like 20 percentage points.”
“Between 1965 and 2010, we find a striking worldwide reduction in Catholic participation in formal services.”
Robert J. Barro
The findings were made possible by an innovative new dataset, mined from the International Social Survey Program (ISSP). By pulling answers to survey questions about religious service attendance in childhood, Barro and his collaborators have compiled the first reliable information on long-running trends in 66 countries. For some places, the numbers stretch as far back as the 1920s.
As a result, the project substantially extends knowledge about fluctuating patterns of religious service attendance worldwide. It also uses an “event study” design to examine two historic episodes: Vatican II and the late 1980s/early ’90s collapse of communism in Eastern Europe.
“This paper is a call to action, to tell people about this extraordinary mountain of data that haven’t been exploited,” said co-author Laurence R. Iannaccone, an economics professor at Chapman University. “It can be used to reconstruct rates of church participation across numerous nations — and we should be studying it very closely.”
“It can be used to reconstruct rates of church participation across numerous nations — and we should be studying it very closely.”
Laurence R. Iannaccone
The first-of-its-kind statistical analysis owes a debt to the late Andrew Greeley, a priest and best-selling romance novelist who earned a sociology Ph.D. in 1962. As a staffer at the University of Chicago’s National Opinion Research Center, Greeley worked for years on the nationally representative General Social Survey. According to Iannaccone, who knew and admired Greeley, the iconoclastic scholar eventually convinced the center to add queries on childhood service attendance. Also asked was how frequently parents had attended while respondents were growing up.
At Greeley’s behest, similar language was added to the internationally representative ISSP in 1991. Iannaccone, director of Chapman’s Institute for the Study of Religion, Economics, and Society, was the first to flag the potential of these retrospective questions in an unpublished paper from 2003.
“I thought, wait a minute — this is like a telephone call across time,” recalled Iannaccone, who suspected those living under oppressive regimes would be more likely to answer honestly about the past.
Barro, co-editor of the Quarterly Journal of Economics, was struck by Iannaccone’s idea and encouraged him to prepare his paper for submission. But life got busy, the Chapman professor said, and he never followed through. Eventually Barro enlisted former Harvard research fellow Edgard Dewitte to help complete the analysis using new ISSP data.
The three co-authors culled more than 200,000 responses covering six continents from surveys conducted in 1991, 1998, 2008, and 2018. “So an 80-year-old who answered the 1998 survey could take you all the way back to the 1920s or ’30s,” Barro explained.
A series of exercises ensured the data’s reliability. For starters, the researchers could compare all four sets of ISSP results for a single year. “You might think the memory gets less accurate as time passes,” said Barro, a macroeconomist who often researches religion in partnership with his wife, economics lecturer Rachel M. McCleary. “But we found it was remarkably stable.”
The information also checked out against contemporaneous results generated by more targeted surveys. Greeley himself co-authored a 1987 paper citing Gallup poll evidence of falling numbers in the pews within the U.S. Retrospective data from the ISSP also stood up against four decades of World Values Survey results from 48 countries.
No quantitative analysis could pinpoint why Vatican II was so alienating. Critics have argued that it undermined Church hierarchy and sowed division between reformers and traditionalists. For his part, Greeley pegged it on Pope Paul VI overruling a special commission’s recommendation to soften the Church’s stance on contraception. “It was this development, more than any other, that shattered the authority structure,” he wrote in 1998.
But the economists can confirm that the trend was unique to Catholicism. They did so by comparing traditionally Catholic countries — where more than half the population identified with the faith in 1900 — to other countries in the sample, including Christian ones dominated by Protestant or Orthodox affiliations. Also studied were individuals with Catholic versus non-Catholic parents.
Monthly religious-service attendance rate across the world
By hemisphere
By region
The trendlines show nothing distinct prior to 1965, when Vatican II’s changes were announced. But then a sudden decline in monthly service attendance emerges among Catholics and within Catholic-majority countries including Ireland, Italy, and Spain. Other events, including a series of sex abuse scandals, likely contributed to the sustained effect the researchers quantified over subsequent decades.
“In a sense, the findings just bolster something people have been saying for a long time,” Iannaccone remarked. “We’re now able to see, across a broad range of countries, that Vatican II had the long-run effect of substantially reducing rates of church attendance.”
A less dramatic finding concerns the fall of the Soviet Union and its allies in the Eastern bloc, all hostile to communal worship. The data offered the economists no hints of resurging attendance once these communist governments started toppling in 1989.
“We were pretty surprised by what we found there,” Barro said. “We had bought into the view that the end of communism promoted a revival of religion.”
The dataset also helped enumerate something the co-authors call “the great religious divergence.” The Global North and Global South were statistical lookalikes in 1950, with an average monthly participation rate hovering near 55 percent. By 2010, rates had fallen to 28 percent across the northern hemisphere while numbers were flat across a series of under-surveyed countries within African and Latin America.
Iannaccone hopes this work inspires a new generation working on population surveys.
“By adding just a few questions, we can learn not only what’s going on today in Portugal or Armenia or the Baltic states,” he said. “We can find out a lot about what happened in the distant past, maybe even in places like China where … the government is wary about asking and the people are even more wary about answering.”
