Research Matters: Eric Posner on “Quadratic Vote Buying as Efficient Corporate Governance”
Research Matters is a biweekly feature in which a member of the faculty talks about some of his or her latest work and its impact and relevance to law and society.
Eric Posner, Kirkland & Ellis Distinguished Service Professor of Law, and E. Glen Weyl, Assistant Professor of Economics at the University of Chicago, wrote “Quadratic Vote Buying as Efficient Corporate Governance,” to be published in the next issue of the University of Chicago Law Review. The authors promote a new voting system for shareholders called Quadratic Vote Buying. They argue that Quadratic Vote Buying would avoid many of the pitfalls of the one-share, one-vote, majority-rule system that is predominant in today’s corporations. Posner spoke about the research.
Q. How did this research come to be?
A. My coauthor, Glen Weyl, invented the idea of Quadratic Vote Buying. In talking with him about it, it occurred to us that it would be a very useful idea in the context of corporate governance. Corporations are governed by shareholders, and shareholders govern by voting, but the voting system shareholders use doesn’t work very well. Quadratic Vote Buying should improve corporate governance.
Q. How does that work?
A. Under the current system, shareholders pay for voting, in effect. You get one vote per share. If you own 1,000 shares you get 1,000 votes, if you have one share you get one vote, and then usually majority rule determines what happens. The way quadratic voting works, or one way it could work, is that if you’re a shareholder you get to buy votes, and you can buy as many votes as you want, regardless of how many shares you own. When you buy a vote, you pay an amount that’s the square of the number of votes that you cast. If you want to cast one vote, it costs a dollar, and if you want to cast 10 votes, it costs $100. If you want to cast 100 votes, it costs $10,000, and so forth. The amount collected is then redistributed back to shareholders pro rata. The basic idea is that when a person votes, what he’s trying to do is influence the outcome of the election in his favor. So if you’re a shareholder and you want the firm to merge, you’re going to vote in favor of a merger because you think it’s going to benefit you. And we want people to vote in order to express their interests and to influence the outcome, but the problem is, when you vote, you’re not taking into account the people who disagree with you, whose interests are the opposite. You’re trying to outvote them. Voting has this mixed quality. It’s both good in allowing people to express their preferences, but it’s bad in that, if enough people express their preferences in one way, they could harm people whose preferences are different.
The genius of quadratic voting is that if you make people pay the square of their votes, you actually force them to balance these two effects. One way to think about that is if you don’t really care about the outcome, but you slightly prefer one outcome to another, you might buy one or two votes. But you’re also not going to affect the outcome very much, because there are all these other people voting. On the other hand, if you care a lot about an outcome, then you should be able to express that by casting a lot of votes. The more votes you cast, the more likely that you’re going to be the pivotal voter who causes the outcome to come out one way, with the result that you could hurt a lot of people. The way to make sure that you only do that when you’re really sure that you’re right, and that your interests are really at stake, is to make the amount of money that you pay increase rapidly, given the number of votes. So if you want to have a huge influence on the outcome, you can, but you have to pay a vast amount of money to do that. That money will be redistributed back to all voters pro rata, at least partially compensating those who are outvoted.
Q. Why is this interesting to you personally?
A. Corporate governance is an important topic, and people are quite unhappy with the current system. So it’s exciting to be able to contribute an idea that could be valuable for something that’s of importance. And in addition, I’ve always been interested in the question of how to distribute public goods which, in this context, would be something like the merger of a corporation. But we’re also writing about other contexts, such as a government’s decision to build a park, or New York City’s bike share program. New York could have used Quadratic Vote Buying to figure out whether people really wanted this bike share program, for example. It’s possible that Quadratic Vote Buying could be used by the government to determine what’s in the public’s interest in all kinds of settings—which is the holy grail of policymaking.
Q. What can change or be influenced by this?
A. At the most general level, corporations are free to change the voting system. And if we’re right, they should and will, or at least some should and will. But a more specific application is to “poison pills,” which are devices that corporations use to prevent hostile takeovers. If an outsider tries to buy a lot of shares, the poison pill automatically makes it cheaper for insiders to buy more shares, so it’s going to be harder to outvote them. Poison pills are generally frowned upon, because people think it’s a way for