Eric Posner and Glen Weyl are professors at the University of Chicago, Mr. Posner in the law school and Mr. Weyl in the department of economics.
Last month, Rupert Murdoch’s News Corporation divided into two entities and inserted shareholder rights plans, or “poison pills,†in their charters to shield them from hostile takeovers.
Shareholder activists are up in arms. The poison pill “is one more tool to entrench management,†one activist told The Financial Times when News Corp. first announced its plan. The problem, however, is not the existence of poison pills but their design: a more carefully designed poison pill would address activists’ concerns while preventing abusive takeovers.
A poison pill deters takeovers typically by giving existing shareholders (including management) the right to buy and vote additional shares at reduced price when a takeover is initiated. It was invented in the 1980s in reaction to a wave of corporate takeovers. Managers disliked takeovers because they frequently lost their jobs in the resulting reorganization, and argued that takeovers disrupted a firm’s operations, caused layoffs, and destroyed shareholder value. Shareholder activists replied that managers used poison pills to entrench their positions at shareholders’ expense.
Because of widespread fears of “barbarians at the gates,†managers ultimately had their way. But given that both sides have advanced reasonable arguments, reform should aim not to eliminate poison pills but to redesign them to address both sets of concerns.
We propose a device that will advance just this goal. In our academic work, we have argued that a voting procedure called “quadratic vote buying†should be used to improve corporate governance. Under this procedure, shareholders pay for their votes, and can cast as many votes as they want at a price equal to the square of the number of votes they cast. The proposal is fairly radical and may take some time to implement. But a more modest version could be applied just to poison pills.
Under our approach, the corporate charter of a firm would provide that when an outside shareholder starts a takeover attempt, existing shareholders would have the right to buy votes for or against the takeover. The price that shareholders must pay is the square of the number of votes they cast, so one vote costs $1, two votes cost $4, and so on. Votes would be aggregated and then resolved by majority rule.
For example, if one shareholder buys four votes in favor of a takeover at cost of $16, and another buys five votes against the takeover at cost of $25, the takeover would be voted down (5-4). The funds would be paid into the corporate treasury, and hence ultimately distributed back to the shareholders albeit on a per-share basis, so if the shareholders each own one share, they receive $20.50 each.
Quadratic vote buying, or Q.V.B., aggregates shareholders’ independent judgments as to the advisability of the merger, taking into account the level of intensity with which they care about the outcome. As a result, the takeover can take place only if it maximizes shareholder value.
Why does this work so well? The answer, as one of us (Glen) has proven in an academic paper, is that Q.V.B. forces the voter to pay a price per vote equal to the number of votes she buys, and therefore gives her an incentive to buy votes in proportion to her marginal benefit from influencing the takeover decision. This is economics jargon but the underlying intuition is easy to understand. Q.V.B., unlike ordinary voting, forces people to pay a lot if they care a lot about the takeover decision, and a little if they care only a little. This enables people with strong views to exert more influence on the outcome than people with weak views, but only if they compensate others at an increasingly high price that is commensurate with their greater influence over the outcome of the vote.
By contrast, if a takeover is put to a regular vote, outsiders can exploit the passivity of most shareholders to obtain a controlling stake and thereby to transfer resources from remaining shareholders to themselves.
Q.V.B. differs in detail rather than in kind from standard poison pills. Under News Corp.’s poison pill, when someone buys 15 percent of the company’s voting shares, existing shareholders have the right to buy stock at half its existing price. Beyond the obvious arbitrariness of a 15 rather than 10 or 20 percent threshold, and half rather than a quarter or two-thirds price, this system has the unfortunate side effect of allowing existing managers to buy shares, and thus votes, more cheaply than the acquirers, and thereby block a takeover that benefits most shareholders.
This is why shareholder activists oppose poison pills. Under Q.V.B., such exploitation would be impossible because the managers would be forced to pay the same quadratic price as anyone else. And yet in the context of poison pills, where voting rules are already adjusted according to the corporate charter in order to fend off takeovers, Q.V.B. seems like a difference in degree rather than kind, and thus ought to be taken seriously by corporate boards.
An even simpler version of could also be used. Under what we have called “square root voting,†or S.R.V., every shareholder may cast a number of votes equal to the square root of the number of shares she owns. S.R.V. is functionally nearly identical to Q.V.B., but avoids the need to redistribute funds, and so is a less radical departure from ordinary corporate governance norms.
Q.V.B. and S.R.V. remove the toxins in poison pills without giving immunity to incompetent management. If management is good for the company, takeovers will be voted down. If not, takeovers will be approved. These devices would be an antidote to many of the woes of corporate governance.
