Guido Imbens: “Receiving a Nobel Prize is a very strange, unreal experience”

On receiving the Nobel Prize

“It’s a very strange, very unreal experience,” Guido said of the moment when he discovered he was to receive a Nobel Prize. “They make the decision around 11am in Stockholm and call you at around 2 in the morning in California. I woke up with the telephone call and you have about half an hour to get ready because you are going to get lots of calls. You realise it’s going to be a very big change. My wife and I woke up the kids, and they got very excited. It was a very special morning.”

From Rotterdam to Hull: a life-changing journey

His return to Hull was perhaps more nostalgic. When he arrived at the train station the memories came flooding back from the time he first arrived in Hull by ferry as a student from Rotterdam. Now, he was coming back as a Professor of Economics at Stanford Graduate School and the recipient of one of the most prestigious awards in the world.

When asked if there was a ‘causal relationship’ between coming to Hull and getting a Nobel Prize, Guido laughed but admitted that it was only after coming to Hull that he thought about doing a PhD in Economics.

“Yes, there’s definitely a big influence there,” he admitted. “But it also got me thinking very differently about research.”

In the 1980s Guido came to the University of Hull as part of an exchange programme with his university in Rotterdam. Thinking it would be as good a place to go as any for a year, he quickly found that the experience was very different to what he was used to.

“When I came here, I really enjoyed the year. It was much more of a campus university than Rotterdam.”

Guido was invited to stay for an extra year as a Research Assistant before being advised by his supervisor and mentor, the late Tony Lancaster, to pursue postgraduate work in the USA.

“He was a very good mentor in terms of letting his students develop their own ideas, but kind of steering them in various directions,” Guido said. “That got me very interested in doing research.”

That early interest went on to develop into work that supported the research efforts of others.

Redefining evidence: the credibility revolution in economics

“The methods that I developed together with Josh Angrist have made some of the research in economics much more credible,” Guido said. “When we started in the mid-80s there was something of a crisis in economics where people felt that a lot of the empirical work that was being done wasn’t credible and people didn’t believe the results. We were at the beginning of a movement that tried to make the work more credible. Angrist later called it a credibility revolution where we developed a bunch of methods that became widely used that changed the way people did empirical work in economics and the way they thought about it.”

Guido humbly dismisses any suggestion that he might have known the significance of the work he was doing at the time.

“I do remember when I was working on it when I was really very happy with it, and I felt that we’d made a lot of progress and understood something that nobody else understood at that time. But I was thinking more at that time like ‘this is a very good idea by my standards’ and that I might be able to publish it in a good journal. That was the extent of my thoughts about it.”

What made this work stand out and lead to its influence being recognised?

“There were a lot of factors and a lot of luck. It was also at a time when the profession was ready for this type of research.”

Why good questions matter

Guido’s work is all about finding the right way to ask and answer questions using real-world data - where experiments aren’t always possible.

“Econometrics is about the study of data,” Guido said. “What I specialise in is trying to figure out methods for drawing causal inferences from observational data. In economics, you typically can’t do experiments. You can’t randomly assign some people to college and keep others out, just to figure out what the causal effect is of going to college on labour market outcomes. We tried to draw those causal inferences from observational data. One of the things I contributed to was developing methods for doing this in settings which are called natural experiments, where there is some idiosyncrasy in the data that helps us uncover the causal effects even though we didn’t do a randomised experiment.”

Bringing econometrics to big tech

Guido’s research is still at the forefront, working with big tech companies such as Google and Meta, all of whom have offices that are cyclable from his office at Stanford.

“They’re doing much more complex experiments than have been traditionally done,” he said. “Nowadays a lot of these companies are doing experiments in settings with multiple populations of interacting agents – maybe buyers and sellers or rental properties and customers. Whatever you do to one group will influence the others, and you must consider that to make sense of the results. Data science and economics come together in a very interesting way.”

Final reflections: asking good questions

We asked Guido what it meant to return to Hull and speak to new graduates. He was modest about offering advice, recognising the role of timing and luck in his own journey. But he returned to the same principle that has defined his career - the quote now immortalised on a fridge magnet in Stockholm:

He said: “The importance of being willing to ask questions and asking good questions. Over time I’ve been much more willing to ask questions, make myself vulnerable and show that I don’t necessarily know all the answers. But asking good questions is important to improve things.”