On Eminent Domain: A Proposal
Eminent domain -- government seizure of land without consent but with payment -- is awful policy. Living nearby to the Long Branch area, which has lost two mayors in the last decade to eminent-domain-related bribery scandals (1, 2), the extrajudicial nature of the process transforms it into a vehicle for corruption.
Even if that were not true, eminent domain would still be deeply problematic. First, its use violates individual rights to property -- see here for more on the Long Branch example. Second, eminent domain as a tool for governments leads to Pareto-inefficient outcomes in the public-development game, in the economists' sense of the word game.
And yet, I do think that state and local governments do need some sort of eminent-domain power on very limited occasions. (This debate between Richard Epstein and Walter Block may be of interest.) If government wants to build a high-speed Acela train which runs from Washington, D.C. to Boston, and there's only one farm in Maryland which blocks the way -- the Acela tracks have to be relatively straight -- I'm not going to sit here and advocate the non-agression principle. I'm going to say that the Maryland homesteader has got to cede ownership, voluntarily or involuntarily.
Nevertheless, there exists a middle range of cases in which the prevailing model of eminent domain fails us. In this post, I want to develop a brief, descriptive game of eminent domain as it now exists and a new model for eminent domain which better protects property rights, creates protections against government corruption, facilitates appropriate public-use development, and leads to Pareto-optimal outcomes for all residents.
When government must build, the decisions around eminent domain are malformed. Furthermore, government's major cost/benefit considerations comes down to legal expenses and its ability to underpay the residents it coerces for their land. This means that governments deciding to use eminent domain rationally tend to apply it in areas where residents have lesser ability to assert property rights in court and/or lesser political power to command a higher payment for their land. In other words, eminent domain encourages governments to take advantage of the poor and powerless.
If eminent domain is to generate Pareto-efficient outcomes, then government and residents need to be making decisions based not upon legal and political expediency but rather upon utility and opportunity cost. In the case that it wants to build a bypass highway, government needs to pick the path which accrues the smallest opportunity cost. That will depend on the value of different parcels of land, the benefits of different paths, and the costs of construction -- not on who's in office, who has the best lawyer, or who shouts loudest at town hall meetings.
Under current practices, if you leave voluntarily, you can get a rather high payout -- but if you hold out, and government chooses eminent domain, then the payout is low. And there is no opportunity cost for residents to turn down bids if they know government will not use eminent domain.
My solution is to make government's bid the vehicle for coercion. Government should place bids for the land it wants. It will do this based on how much government wants to pay. If the residents take the bid, the process is over. If they turn it down, the value of their property is re-assessed immediately at government's offer, and their property taxes are prorated for the year based on the timing of government's bid.
If the residents continue to sit on the land, then they have made a decision based on the utility of their land after the elevated taxes as compared to the offered payout. If government decides it really does need that particular plot of land, then its value has increased to the government. It will offer a higher bid, and when it does, the property tax reassessment will happen automatically.
Imagine we get to a point where government is considering a bid of $10 million for a plot of land whose fair value is one percent of that, had there been no highway. Why is government bidding so much? Because it needs the property and -- here's the key -- it is being forced to consider the opportunity costs of the resident, such that it can bid the minimum it will take for the resident to sell their land. Why is the resident holding out for so much? Because he or she has decided that the property is worth that amount to them, and they are willing to "put their money where their mouth is" as to their value assessment when they have to pay exorbitant property taxes.
By linking government's bids with property taxes, my proposal can avoid a considerable amount of takings by eminent domain and re-route them through free-market, Pareto-efficient processes.
Note: Further thinking about this today suggests that holdout problems could still be severe. The final farm owner, for example, blocking the construction of a bypass can command large economic rents if he or she is willing to pay a few years of taxes in hope that will twist the arm of the government into increasing its bid. This may provide an inappropriate incentive for people to hold out. The solution may be to impose a cut-off, say 80 percent, of landowners by acreage in an eminent-domain zone whose consent is required and then to impose the eminent-domain coercion on the remaining 20 percent, compensating them according to the payout received by the 80 percent. This, however, raises philosophical questions about the appropriate threshold -- why not 51 percent, or 1 percent? -- and the rights of a holdout minority.