In a discussion of mechanism design theory, an economics professor from Harvard explained that the key to designing a mechanism for fair decisions is that the mechanism must be incentive compatible.
Dr. Eric Maskin is the author of more than 100 papers on economics and won the Nobel Prize in economics in 2007. After what he referred to as a “pleasant and stimulating day” at Linfield, he presented real-life examples in his lecture on mechanism design.
His first example was simple. A mother wants to divide a cake between two children. Her goal is that each child will be happy with their share of the cake. This means that both children, Bob and Alice, must think that they have half of the cake.
If the mother knows they see the cake in the same way that she does, she could divide it equally from her point of view. But what if Bob sees the cake in a different way than the mother does?
The mother wants a fair division, but the problem is that she does not know how the kids see the cake. “In effect, she doesn’t know what’s fair,” Maskin said.
How can she design a fair division if she doesn’t know what’s fair?
This is the problem that mechanism design faces and attempts to solve.
“This is a very old problem that has been around for thousands of years,” Maskin said.
He explained that the traditional solution would be to let Bob divide the cake in two and let Alice choose the half she wants.
This way, Bob has incentive to divide equally. Bob wants to secure his happiness, and Alice will be happy because she gets to choose.
Maskin explained that the key features of mechanism design are that the designer does not know the optimal outcome, so they must proceed indirectly by having the participants generate information to identify the optimal outcome.
However, another problem arises because the participants do not care about the mechanism designer’s goals. This is why the mechanism must be incentive compatible.
The goal of the government is to put the transmit license in the hands of the company who values it most, which is called an “efficient outcome.” The problem is that the government does not know who values it most.
The government could simply ask who values it most, but many companies have incentive to exaggerate. This way, there is no guarantee that the license will go to the right company.
The solution, he explained, is to have every company make a bid. They then award the license to the highest bidder, who must pay the amount of the second highest bid.
This way, they have no incentive to understate. They are forced to bid what it is worth to them, which is what the government wants.
The goals of the mechanism designer have been achieved by allowing participants to be involved while eliminating possible problems.
He explained that mechanism design prompts the question: is there a process that we can follow that will answer whether a given goal is viable? Can we find a mechanism that works?
This is the issue that mechanism design theory addresses, and he explained that they can generate a process for finding this mechanism.
“Don’t worry. I won’t show you why,” Maskin said, jokingly.
He instructed the audience to see his paper “Nash Equilibrium and Welfare Optimality,” which explores the topic further and explains how this process works.
Samantha Nixon/Staff writer