Bringing people together for the best possible outcome is the foundation of marketplaces like eBay. And the concept doesn’t just apply to buying and selling; it can also be used to solve complex problems like matching organs with patients who need transplants and matching medical residents with hospitals. On Tuesday, Nobel Prize winner and Stanford economics professor Alvin Roth shared insights from his groundbreaking work on computer-assisted markets as part of eBay Data Labs distinguished speakers program.
In the 1980s, Roth began building on fellow economist Lloyd Shapley’s work on cooperative game theory to explain how markets function in practice. According to the citation for his 2012 Nobel Prize in economics, which he shared with Shapley, Roth used experiments and empirical studies to show that stability in matching participants was critical to successful marketplaces.
“It should sound a lot like marriage. In matching markets, you don’t only choose. You also have to be chosen.”
One area where Roth tested his ideas on markets was the field of organ transplants.
The first kidney exchange in the U.S. – involving a pair of donors and recipients in simultaneous transplant surgeries – was done in 2000 in Rhode Island not long after Roth went to work at Harvard University, and it was organized by the patients’ spouses, who met while sitting in a waiting room at a dialysis clinic.
“As an economist, you say, ‘That’s no way to design an exchange – we can do better than that,’ ” Roth said.
He was drawn to study kidney exchanges because of the mathematics surrounding them and was inspired by an economics problem he first heard posed back in the mid-1970s: How would you trade houses if you couldn’t use money?
Roth realized that theoretical problem could have real-world implications for kidney transplants, which by U.S. law can’t involve money.
Transplants – in addition to being hampered by greater demand for organs than supply – can be cumbersome for many reasons surrounding the compatibility of living-donor organs and patients. For example, a person willing to donate a kidney to a relative or friend in need isn’t always a match on both blood and tissue type. But if the donor were still willing to give up a kidney to ensure their friend or relative received one from a compatible donor in the same situation, chains of transplants – efficient exchanges – would be possible and more lives saved.
“You can’t say, ‘Give me a kidney today, and I’ll give you one tomorrow,’ ” Roth said. “The trouble with barter is you need a double coincidence of want: You have to find someone who wants what you have and has what you want.”
To help increase efficiency in transplants, he designed an algorithm to match available donor organs and recipients without a limit to the length of cycles, a development that has led to longer chains. In December 2011, computer matching based on his work led to a kidney exchange known as Chain 124, a 60-person chain of transplants – the longest ever.
Because of the complexity and constraints of such an exchange, “it’s something you could only do with a computerized market.”
And the larger lesson from the evolution of kidney exchanges? “When you make a market, people start to change their behavior and the marketplace changes.”
That’s something Roth has seen from the beginning at eBay, on which he wrote one of the earliest academic research papers, focused on user behavior.
“eBay is, of course, an iconic company, one of the first computer-assisted markets,” and one that also relies on reputation systems and trust between buyers and sellers, he said. The company’s commitment to continually improving the user experience is important “because the data aren’t perfect. You often have to re-optimize.”
After the talk, employees surrounded Roth to ask more questions about his work, and that’s exactly what eBay senior research director Neel Sundaresan, who organized the event, hoped would happen.
“Bringing in a person like Al Roth, who is a great achiever and an amazing storyteller, exposes employees’ minds to areas not directly related to their work but has applications way beyond what they do in their workday,” he said. “Al made complex concepts look so simple and tied a number of examples back to eBay – matching marketplaces and search theory – and kept the audience engaged. I feel fortunate that we at eBay Data Labs could host a Nobel laureate like Al Roth and learn from him and provide this opportunity to others as well.”
Others who got the opportunity to hear from Roth were high school students from Tennyson High in Hayward, Calif., a partner school for eBay’s Inspire! Program. The program aims to equip students with the tools they need to build lasting careers through STEM education.
Roth urged students to learn new things, keep an open mind, take advantage of the college environment, and discover where their passions lie.
“With friends, dorm mates and colleagues, you will have a number of opportunities to get into interest groups that you didn’t have before.”
Recalling own education, Roth shared how he was bored in high school and dropped out, but he later enjoyed college very much.
Sundaresan said he knew the students benefitted from Roth’s encouragement.
“It’s important for them to learn that such high achievers do not necessarily have a great start and they still do amazing things in life. At an age where the students are nervous about their education and their future, such touch points give them the comfort that if one keeps an open mind and is sincere about one’s pursuit, one can achieve greatness.”
That lesson really resonated with the Tennyson High students.
“Alvin Roth has become my new motivation to continue to succeed considering he didn’t graduate from high school but he went on to win the Nobel Prize in economics. This helped me realize that even if you ‘fail’ in high school, college can be a new start,” said junior Rajneet Sandhu.
Senior Marlene Rodriguez added, “It was very inspiring to me to see and learn from the eBay environment. Meeting an actual Nobel Prize winner will definitely be an experience I will always remember.”