Thought Leadership At UC Davis Graduate School of Management: Professor Hemant Bhargava On Technology-Based Business & Markets

Facebook. Google. Uber. Netflix. Tesla.

What do they all share in common? They are highly successful and disruptive businesses based on platform economics. It’s a topic that Professor Hemant K. Bhargava knows inside out.

An academic leader in economic modeling and analysis of technology-based business and markets, he has studied platform businesses in the information and telecom industries, healthcare, media and entertainment, and electric vehicles.

Bhargava’s academic journey began at the University of Delphi in 1981 where he majored in mathematics. He earned his MBA from the Indian Institute of Management in Bangalore in 1986 and eventually picked up his PhD in information systems, operations, and economics at the University of Pennsylvania’s Wharton School in 1989. He has taught at both Penn State University and the Naval Postgraduate School.

Yet for the past 21 years, Bhargava has made the University of California at Davis his home as a successful scholar and teacher and academic innovator. Among other things, as an associate dean at UC Davis Graduate School of Management, he has overseen the MBA programs and currently serves as director of the school’s Center for Analytics and Technology in Society. Bhargava conceived and designed and also leads the school’s Master’s in Business Analytics program.

In this wide-ranging conversation, part of the Thought Leadership Series at UC Davis Graduate School of Management, Bhargava looks back on his career and the insights he has developed from his studies on business platforms. He is interviewed by former Businessweek Executive Editor and Poets&Quants Editor-in-Chief John A. Byrne.

John A. Byrne: You’ve had quite an academic journey starting with your undergraduate degree from the University of New Delhi to your PhD in decision sciences from The Wharton School. And you have been at UC Davis for two decades. What inspired you to become an academic in the first place?

Hemant Bhargava: I studied math in college and I was always very excited with math. And then I went to business school and started seeing applications of math in business. But this was in the 1980s in India and we had just started seeing computers and computing. And I got my first experience with computing at the business school and it fascinated me. And I remember reading books at that time on artificial intelligence. I was just amazed at some of the things that were going on in AI at the time.

So the journey to being an academic was sort of partly intentional and partly accidental. I went to one of the top mathematics programs, but I often found myself in front of the classroom trying to re-explain things that were being explained by the instructor. And then at the business school, which was also the top program in the country with great professors, I found myself often skipping lectures and not reading the prescribed books. Instead, I discovered my own material and was often in the library. So I think that intellectual curiosity and desire to explain complex things to other people connects to being an academic. 

That’s the intentional part. The accidental was then I got a job at one of the really great IT startups in India. I was very happy, but I had a sister in the U.S. who was doing a PhD in management science who insisted I should consider coming to the U.S. And she ended up calling people at The Wharton School after I’d taken my GRE. So I had not applied anywhere, but then I got a call from Wharton asking me to apply and that they wanted to admit me with a full scholarship. That sounded like a good offer and that’s how I got there. But I still had not really thought what it would mean to be an academic. Once you’re in a PhD program at a place like that, the most frequent and expected path is to be an academic.

Byrne: That totally makes sense to me.

Bhargava: And I love both the research and the teaching. That led to my getting into the academic world, and that’s where I went to the Naval Postgraduate School in Monterey. And it was interesting to have students who were older than me when I started my job, but great students from the military. And some really wonderful opportunities to work with the Department of Defense and other agencies on real problems.

Byrne: Your interest in your topic area can be traced to your early interest in mathematics. How did your research interests evolve over the years of your career?

Bhargava: When I went to Wharton, I was interested in computing and technology but also math, which at Wharton was called Decision Sciences. So it was the science of decision-making. And I got involved with a project that was sponsored by the U.S. Coast Guard to build decision support systems for them. And my Wharton, Steve Kimbrough, who was leading the project. He used to mention this quote from one of the officers at the Coast Guard who was behind the project: ”Don’t just give me answers, but tell me what’s behind all of this.”

Byrne: Right.

Bhargava: And that really drove a lot of our thinking. This was in the late 1980s, before the internet had happened, but we created something we called generalized hypertext where a machine produces the links. So rather than sit and make links to things, and this was driven by the idea of where did the numbers come from? So you could click on a result and it would trace you back to the models, the equations, or even the parameter values that were entered by someone. If I look back at it now, 35 years later, it was really a cool innovation at the time. So that sort of got me into computing in the context of decision making and business modeling.

Byrne: Now you have over 100 peer reviewed articles published. I wonder which one of them landed with the greatest bang.

Bhargava: There’s an easy answer and a complex answer, as always. The easy answer would be a paper we worked on in the early 2000s when Internet search engines were happening. And this was led by my doctoral student at the time, Joan Fung, when I was at Penn State. She went for a summer internship at Yahoo Research Labs, and we ended up writing a paper comparing methods for rank and search results with advertising. That paper was published in 2007, takes a long time, and in 2019 it won a Test of Time Award because after a dozen years or so, it had shown its value through a number of methods and citations.

Byrne: And you discovered in the paper that there is a correlation between whether a firm advertises with a search engine and how high its results organically come up.

Bhargava: That’s right. That’s sort of the premise part of the work. But what we showed was that when you’re ranking results, the practice then was to rank based on payment alone. And another emerging alternative was to rank based on a combination of payment and perceived content relevance. So the first approach was, “Look, we want to make the most money through these ads, so let’s rank whoever pays the most and put them on the top.” And we showed that in many situations that combining payment and potential relevance was a better way to rank results.

But then we also showed something really interesting. If you have new pages that are asking to be ranked, you know nothing about them so then you need to try them out. And we demonstrated and analyzed different methods for doing these trials so that you could explore new content while also exploiting the historical relevance that they had done. So that’s the easy answer.

Byrne: What’s the more complicated answer?

Bhargava: The more interesting and complex answer is actually going back to that work on generalized hypertext. A few years later, my co-colleague Ramayya Krishnan, who’s at Carnegie Mellon University, and I decided to do something in the context of decision modeling, but also internet computing. And this is the late 1990s when the internet was just taking off.

We ended up building DecisionNet, meant to be an electronic marketplace of decision models and data and users. We were interested in decision modeling, and that led to my research on platform economics. We didn’t make a big deal of it at the time, but then five or six years later, my research transitioned to the economics of technology and platforms and I’m doing that now and it’s led to some really cool papers. But I think that journey is really interesting from the hypertext to this DecisionNet electronic marketplace. We’ve looked at how to model digital platform businesses like Uber and Facebook and so forth. And the latest thing I’m doing now is looking at digital addiction, which is another side effect of the advertising driven revenue models that a lot of these platforms adopt.