SIEPR Senior Fellow Guido Imbens wins Nobel in economic sciences

During his first year teaching and living at Harvard University, Guido Imbens spent his Saturday mornings with his colleague Joshua Angrist at the university housing laundromat, and during the spin cycle, the pair would discuss work and ponder questions about the world. Memories of those leisurely weekend conversations, nearly three decades ago now, flooded back to Imbens today upon learning that both he and Angrist have been awarded the 2021 Nobel Prize in Economic Sciences for work they conducted together.

“We were talking about ideas and thinking about open questions and it’s just kind of amazing to think that that work just very directly … it’s essentially where we figured out the main ideas that are the basis for this prize,” said Imbens.

Imbens is the applied econometrics professor and professor of economics at the Stanford Graduate School of Business. He is also a professor of economics in the School of Humanities and Sciences and a senior fellow at the Stanford Institute for Economic Policy Research (SIEPR).

The Royal Swedish Academy of Sciences, which grants the Nobel prizes, jointly awarded one half of the $10 million Swedish kronor (about $1 million USD) prize, officially known as the “Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel,” to Imbens and Angrist for “their methodological contributions to the analysis of causal relationships.” The other half of the prize was awarded to David Card at the University of California, Berkeley, for “his empirical contributions to labour economics.”

“On behalf of the Stanford community, I would like to offer my warm congratulations to Guido Imbens on this remarkable achievement,” said Stanford President Marc Tessier-Lavigne. “Professor Imbens has played a central role in shaping how researchers understand and analyze causal relationships. His insights and framework, developed alongside his close colleague and fellow laureate Joshua Angrist, have helped researchers in the field of economics and beyond use natural experiments to answer important questions for society and our world. All of us at Stanford take great pride in his accomplishments.”

Throughout his career, Imbens has conducted influential work to help address the limitations of real-world experiments of social scientists, greatly improving researchers’ ability to assess the effects of interventions from both field and experimental data. His work is used to analyze complicated research questions such as the effectiveness of a new drug on a patient or the impacts of new regulation on economic activity.

“I was absolutely stunned to get a telephone call,” said Imbens over the telephone with the Royal Swedish Academy of Science during a question and answer session they hosted with the media immediately following the announcement. “I was absolutely thrilled to hear the news, in particular hearing that I got to share this with Josh Angrist and David Card, who are both very good friends of mine.”

Mark Duggan, the Trione Director of SIEPR, congratulated his “brilliant and inspirational colleague on today's incredibly well-deserved recognition of the Nobel Prize.” 

Imbens was a supporting influence for Duggan when he was considering whether to pursue a PhD in economics more than 27 years ago. 

“He personifies greatness as a researcher and human being who inspires his students, colleagues, and co-authors every day to pursue rigorous research that accurately isolates causal relationships in economics and related fields,” Duggan said. “Congratulations also to Joshua Angrist and David Card for their very well-deserved receipt of the Economics Nobel this year too for their extremely important research contributions over the years as well!”

Expanding the field of econometrics

During those weekend mornings at the Harvard laundromat, Imbens and Angrist would discuss research and the latest economic papers they had read.

At the time, they both were assistant professors at Harvard (Angrist is now an economist at MIT) and as Imbens recalled during the virtual press conference Stanford hosted Monday morning, “we hit it off right away.”

The pair eventually pioneered a model they called the local average treatment effect (LATE), introduced in a 1994 Econometrica paper, called “Identification and Estimation of Local Average Treatment Effects,” that showed researchers how to draw causal inferences from observational data.

The pair looked at how natural experiments – using real-world situations where chance and randomization naturally occur, as opposed to controlled conditions, which can be costly, time-consuming and in some instances, unethical – can serve as powerful tools for evaluating how cause and effect might be connected.

The model has had a significant impact on research practices in econometrics and other areas of statistics, with some of Imbens’ papers being placed among the most cited economics research of the 1990s. Imbens summarized some of his work in a 2015 book he co-authored with Donald B. Rubin, called Causal Inference for Statistics, Social, and Biomedical Sciences (Cambridge University Press).

By drawing on natural experiments, the LATE model helped Imbens and others examine causal phenomena and address important issues facing the public and policymakers. For example, in order to study the effects of unearned income on labor supply, Imbens looked at evidence from an actual lottery to determine whether unearned income changed incentives for people to work.

With Donald Rubin of Harvard and Bruce Sacerdote of Dartmouth, Imbens surveyed people who played the lottery in Massachusetts, where winners were awarded their prize incrementally over many years instead of in one lump sum. As Imbens explained, this was very similar to what a guaranteed basic income would look like. By studying people who won the lottery and those who didn’t, the scholars could infer the causal effects of having guaranteed income.

What they found was while there was some effect on labor supply, it didn’t change how much people worked. “Without having done an experiment, we could still learn a lot about the causal effects of labor supply in that case,” Imbens said.

This type of statistical modeling can help inform policy leaders who might want to study the effects of federal or state assistance programs, explained Imbens in an interview with Stanford News Service.

“We use that as an instrument to estimate the effect of unearned income on labor supply, which is a very important part of designing social security systems,” Imbens said. “It really showed how you could get credible results with these methods in realistic settings.”

In another seminal paper, Angrist, along with the late Alan Krueger, employed the LATE model to examine the effects of education levels on income. They found that students who were required by educational compulsory laws to stay in school a year longer had higher earnings.

During Monday’s virtual press conference, Imbens acknowledged the tragic death of his collaborator.

“Alan Krueger, who is the only person in the world who actually co-authored papers with David Card, Josh Angrist and myself, passed away a few years ago,” Imbens said. It was “very untimely. We all greatly miss him and wish he could have been here to share this with us.”

Pioneering a new methodology

Imbens’ work has helped inform governments and policy institutions on designing and evaluating economic policy interventions in areas such as education and labor.

“It’s been great to see how the profession has gotten interested in these methods,” Imbens said.

But it took time for Imbens and Angrist’s new approach for addressing societal issues to gain notice. When they began their work in the early 1990s, their unconventional methods generated little attention. Imbens now thinks that might have been a blessing in disguise. “I suppose that was an incredible time for us because we had the room and the space to actually think deeply about these things without having people look over our shoulders,” Imbens said.

He said that there were times it was “somewhat lonely” but having Angrist as a colleague, and others in his former department at Harvard, including Rubin and the late Gary Chamberlain, helped make it an incredibly supportive environment.

Imbens is good friends with both of his fellow Nobel recipients. Angrist served as the best man at Imbens’ wedding to Susan Athey, who is also an economist at Stanford. And when Imbens was teaching at UC Berkeley in the early 2000s, he and Card would often work late in the office and then after they were finished working, they would grab a drink together. During these late-night conversations, Imbens said that he learned a great deal from Card about how to do research and what kinds of questions he worked on.

“To be sharing this prize with both of them is immensely gratifying,” Imbens said.

Helping other researchers learn from their data

Since publishing his groundbreaking paper with Angrist nearly 30 years ago, Imbens has seen an increasing number of students flock to the subdiscipline.

One of those students is Michael Pollmann, a current PhD student in economics. He recalled a comment Imbens made that has stuck with him.

“He said that when there is something in our research that we don’t yet understand, that is the most interesting and greatest part. It reveals something we can still learn and where we can also help others to understand these questions better,” said Pollmann, who is currently studying how to take the spatial nature of data into consideration.

Imbens, as a researcher but also as a mentor, has shown others, including his students how they can use the data they have to answer existing questions, he added.

“He takes this two-step approach of learning what we still need to learn and then, once you know what you still need to learn, figuring out how to actually do it,” said Pollmann.

While much of Imbens’ scholarship focuses on theoretical research and methodology, he stresses the importance of having an empirical understanding of the world – something he also tries to instill in his pupils.

“I try to convey to my students that the way to do good methodological research is to really talk to people who are doing empirical work and policy-relevant things so you can make sure that the methodological work you are doing is going to be used and is relevant for people analyzing data,” Imbens said.

‘He won’

A highlight for Imbens was being able to celebrate the accomplishments with his family. It had already been an exciting few days in the Imbens-Athey household: just last week, Athey, who became the first woman to win the prestigious John Bates Clark Medal in 2007, was voted president-elect of the American Economic Association.

The latest news – delivered in the early hours of Monday morning given the time difference between California and Stockholm, Sweden where the prizes are decided – woke the entire household.

Imbens’ teenage son described the scene in their home this morning after receiving the trans-Atlantic phone call informing them of their father’s Nobel Prize.

“I woke up around 2:30 in the morning. Everybody was running around,” Carleton Imbens, 17, said. “My mom ran into my room and quickly told me the news. At this point, I wasn’t really sure what the day was going to look like so I thought I just needed to get back to sleep because I have school in the morning. But it’s become quickly apparent that that’s not going to happen.”

Athey, Imbens’ wife, later told their three children, “You can decide if you want to go to school or not.”

Imbens’ children have absorbed a fair amount of knowledge about their parents’ work from the family’s dinner conversations. “Mostly they want to talk about other kinds of science, but occasionally they’ll indulge their parents with econometrics,” said Athey, the Economics of Technology Professor at the Stanford Graduate School of Business and a senior fellow at SIEPR.

So Imbens’ daughter knew immediately what her mother was talking about when she opened the door to her room and said simply, “He won.” “We’ve been kind of betting on the Nobel Prize. We didn’t actually think my dad was going to win,” said Sylvia Imbens, 10.

Despite their abrupt awakening, the children quickly rallied, Athey said. “My oldest son was running around getting an extension cord and my youngest opened the door when the doorbell rang and wanted to get everyone coffee.”

Sylvia Imbens said that while the details of her father’s research are still fuzzy to her, what is clearly apparent is the amount of time he puts into his work. “I know how much effort he puts in, so it feels very satisfying for him to get recognition,” she said.

Athey said that it was very sweet to see how proud the kids were of their father, whom they call “Pap.” “He’s a really great dad. He spends so much time with the kids so it was really fun to see how excited they were,” Athey said.

Originally from the Netherlands, Imbens attended Erasmus University Rotterdam and the University of Hull in England.

He received his master’s and PhD degrees in economics from Brown University in 1991. He was an assistant professor and later an associate professor at Harvard between 1990 and 1997, and he taught again there from 2006 to 2012, after which he joined the faculty at Stanford University. He was a professor at the University of California at Los Angeles from 1997 to 2001 and at the University of California at Berkeley from 2002 to 2006. He holds an honorary doctorate from the University of St. Gallen.

Imbens is a fellow of the Econometric Society and the American Academy of Arts and Sciences; the Royal Holland Society of Sciences and Humanities; the Royal Netherlands Academy of Sciences; and the American Statistical Association.