Quadratic Voting as an Input to Cost-Benefit Analysis

When administrative agencies regulate, they are legally required to quantify the costs and benefits of their regulations. Yet most agencies struggle at this task. This is in part because a large number of regulations provide benefits that are not traded in markets and cannot be easily priced. These types of benefits are difficult to assess in monetary terms, even though they are almost surely sizeable. Agencies typically try to price non-market benefits using contingent valuation studies, which are surveys that ask people how much they would be willing to pay without any real money actually changing hands. Unsurprisingly, contingent valuation surveys have proven to be inaccurate and unreliable. Instead, agencies should use quadratic voting (QV) to price nonmarket goods. Quadratic voting is a decision procedure in which voters use actual dollars to buy votes for or against a ballot proposition or candidate. Both the marginal cost of buying an additional vote and the marginal benefit of doing so — the probability of casting the pivotal vote — increase linearly with the number of votes cast. When marginal costs and marginal benefits are equal, individuals are likely to buy votes in proportion to their actual preferences. This leads to socially efficient outcomes. Quadratic voting is particularly suited to administrative regulation because agencies already have the legal authority to use quadratic votes as inputs to the regulatory process. Given the advantages of quadratic voting, and the fact that agencies could adopt QV without waiting for Congress, there is little reason for them not to act