Who do you call to mend a market, or to design a new one?
Top brains at AGORA are creating solutions for complex practical problems
In 2008, the US Federal Communications Commission held a historic 700 MHz spectrum auction that raised more than US$19 billion for the government.
For the first time, new players such as Google were allowed to bid alongside incumbents such as Verizon and AT&T for a valuable spectrum pie that had previously been the exclusive domain of broadcast TV.
The auction’s success was due in great part to new rules co-designed by then professor of economics at Caltech, Jacob Goeree, which allowed the market to discover the best allocation of spectrum. That model has since played a key role in the design of spectrum auctions around the world.
Goeree, a world-leading market designer, is now a scientia professor at UNSW Business School and director of its AGORA Centre for Market Design.
AGORA draws on the expertise of Australian and international researchers from a wide range of disciplines – including game theory, experimental economics, computer science and operations research – to solve major market problems.
The centre tackles real-word issues such as how to design pollution permits for greenhouse gas initiatives, alternatives to standard voting, and how best to trade water rights.
In theory, market design could be used for anything from energy markets to allocating spaces in childcare centres, says Goeree.
“But it turns out the details of a market’s design really matter. That is where the science comes in,” he says. “You also have to face realism and practical considerations. This drives the design of any new market; you can’t escape it.”
'Five or 10 years ago we could not have solved the problem of how to trade so many different types of permits'
MARTIN BICHLER
Advances in computer science
One of the most complicated projects Goeree has worked on is a trading exchange for fishing rights in NSW. The state had more than 100 types of fishing permits defined by species, region and method of fishing.
The exchange had to cater for fishers who wanted to leave the industry, and others who wanted to buy enough permits to sustain their operations.
“But how and where could the fishers meet to trade the rights? Who would they call if they wanted to buy or sell permits? By conducting the exchange over the internet, it made it easier to connect all these different fishers,” says Goeree.
One of the first people he called for help was Martin Bichler, computer scientist and professor in the department of informatics at the Technical University of Munich. Bichler and Goeree had co-authored an influential handbook on spectrum auction design.
Bichler, now an AGORA fellow, specialises in operations research and optimisation. He was attracted by the complexity of the fisheries project and by AGORA’s international, inter-disciplinary team. Advances in computer science also meant it was the right time to solve such a complex market problem.
“Five or 10 years ago we could not have solved the problem of how to trade so many different types of permits,” says Bichler.
“But the type of optimisation and algorithms that have been developed in the past 20 years, in particular in operations research, have been an enabler for these types of markets.”
The research behind the exchange was published in the Proceedings of the National Academy of Sciences and has won three high-profile awards.
Fixing the problem
Goeree and Bichler are now working on several theoretical issues that emerged from the project, and are thinking about how their exchange design could be applied to other markets such as tenders for transportation and logistics. In fact, any market where a customer wants to buy a bundle of goods, says Bichler.
Markets not based on price, and which are not governed by traditional economic theory, can also benefit from market design.
AGORA fellow and professor of economics at Stanford University, Alvin Roth, shared the Nobel Memorial Prize in Economic Sciences in 2012 after pioneering an innovative ‘kidney swap’ program in the US that solved the problem of how to encourage more people to donate kidneys to strangers.
The program linked people who wanted to donate a kidney to a family member but who they were not compatible with, to other incompatible donor pairs. The larger the chain of donors, the greater the chance of multiple matches.
For so long, economists have been theorising from their proverbial armchairs about how markets should work, says Goeree.
“Now we get to design entirely new markets. Or, if there is a failure in an existing one, market designers are like doctors who can fix the problem.”