CAMBRIDGE, MA – Samantha Robbins, an assistant professor of economics at Harvard, has found that Hell has higher gross domestic product (GDP) than does Heaven. Her research uses cutting-edge statistical techniques to support her conclusion that Hell’s institutions have propelled fast economic growth.

“Lots of people have investigated why some places are rich and others are poor,” Robbins told the Ichthus, “But they’ve only focused on our mortal lives. Since religious thinkers have been emphasizing for millennia that we live eternally after death, either in Heaven or in Hell, I thought this was an important new research avenue to explore.”

Robbins says she acquired data from the statistics departments of Heaven and Hell to determine production measures. After a year of research, she found that the key to this difference in GDP was Hell’s institutions.1

“Cultural differences were one possible explanation for the difference in GDP between Heaven and Hell,” Robbins says. “Geography was another, since the lack of a sun in heaven might harm agricultural productivity.2 But I found that, in fact, Hell’s institutions – generally understood as government policies and other ‘rules of the game’ – are driving the higher GDP in Hell.”

Robbins presented her results recently at a meeting of the American Economic Association. “Samantha gave an excellent presentation,” says Bert Johnson, a senior economist at the International Monetary Fund who was present at the meetings, “But some of us there expressed concern that her results were plagued by an endogeneity problem. According to some strands of Christian theology, at least, Hell is a consequence of sin and is endogenous to individual characteristics. For economists, this sort of problem is a big issue. If the kinds of people that go to Heaven and those that go to Hell are different, then even if a Hell-income correlation exists, it may not be causal.”

Robbins says she left the AEA conference with a renewed vigor to establish empirically that Hell’s institutions had a positive effect on income.

“I received many helpful comments during my seminar presentation,” she said, “And then I decided to try to move beyond my rather simple analytic approach and instead decided to try something more advanced, called instrumental variables. For this to work, I needed to find something that would predict which people were in Hell.”

Victor Li, a research assistant who worked with Robbins on the project, explained that they came up with their instrumental variable because of something they’d both seen on the way to work one morning. “We’d heard from a street evangelist by the T stop in Harvard Square that only people who believe in Jesus could go to Heaven,” said Li. “So we decided to use exposure to the Christian message, via missionaries, as an instrument for getting to Heaven.”

This had the promise of being a perfect instrumental variables approach for estimating the effect of Heaven and Hell on income, Robbins said, but problems soon emerged. “Unfortunately, it quickly began to seem as though the association between missionary exposure and getting to Heaven wasn’t that strong,” Robbins said. “Some of the people who had never heard of Christianity, in fact, ended up in Heaven!”

Robbins said that when she’d started to discuss her confusing results with colleagues, news of her research and her most recent findings started to reach the general public. “I especially started hearing from lots of ‘inclusivist’ religious folk who thought that one’s religion didn’t really matter for going to Heaven or not,” she said.

Li said other aspects of the dataset were confusing. “We also found that some of the professing Christians who had been exposed to missionaries didn’t end up in Heaven,” he said.

Robbins explained, “We actually had a variable in the dataset that had extra explanatory notes for the observations, and for those individuals, the variable said, ‘I never knew you! Depart from me, you workers of lawlessness.’ We didn’t understand that at all at first, but somebody looked it up, and apparently these are Jesus’ words in Matthew 7:23.”

Li explained that these results actually turned out to be a false alarm. “Samantha was distressed about losing our promising instrument, so she revisited the code that I’d written to produce these results and realized that I’d done a really poor job cleaning the dataset, and this meant the results thus far had been wrong,” he said. “I actually got fired after that.”

Robbins said that while she was trying to find another research assistant her research agenda was receiving heavy criticism from many people outside the economics profession.

“I received numerous emails from people who assured me that any notion of an afterlife was a crude, pre-scientific idea that had no place in modern economic analyses,” Robbins said. “Others criticized the project on the grounds that no one went to any sort of Hell, and others thought that Hell was just a way of describing the annihilation of people who would later have no income at all. They were quite skeptical of my data.”

Mary DiMartini, a professor of New Testament at Harvard Divinity School, said she also criticized Robbins’ project.

“I was concerned that Samantha’s focus on the afterlife would obscure the real message of the Christian religion, which actually is about fighting oppressive power structures in the here-and-now,” DiMartini said.

Daunted by too many challenges, Robbins says she’s dropped this research project for now and has set out to pursue something more traditional.

“I’m now looking at the relationship between exchange rates and foreign direct investment,” Robbins said. “Religion is just too messy.”

Peter Hickman ’16 is an Applied Math concentrator in Leverett House. He is editor-in-chief emeritus of the Ichthus.

  1.  For technically-minded readers: she used a regression discontinuity approach, focusing on an arbitrary boundary between Hell and Purgatory as a cutoff.
  2. Revelation 22:5: “… and they will not have need of the light of a lamp nor the light of the sun, because the Lord God will illumine them” (NIV).