One of the thorniest problems faced by any state or local government is how to fund its common infrastructure for such essential services as roads. In general, two possible sources of revenue may be tapped. The first are general tax revenues imposed on real estate in proportion to its assessed value. The second source are exactions, or impact fees, imposed on landowners seeking a building permit for a new home. The hard challenge is to determine the mix of revenues from these two different sources.

This issue has puzzled and divided the US Supreme Court, which just heard oral argument in Sheetz v. County of El Dorado—a county with a population of around 200,000 near Sacramento. George Sheetz of Placerville, a recent retiree, had applied for a building permit within the town. It is well understood that these permits can be difficult to acquire but are worth tens or even hundreds of thousands of dollars once granted. In this instance, the county offered the permit with a giant catch: you must pay us over $23,000 to get that permit. The county claimed that the fee was needed to offset the increased costs of building out the highway system to accommodate the extra traffic. That fee was determined under a general legislative formula that used standard, not individualized, determinations that a new home would impose on the system. The county claimed that it would be too costly to set these fees on a retail basis.

Earlier takings cases involving individualized determinations exactions held that constitutionally they had to satisfy twin conditions of “essential nexus” and “rough proportionality,” both derived from Nollan v. California Coastal Commission (1987) and Dolan v. City of Tigard (1994), which both dealt with individual cases. In Sheetz, the county first insisted that the generalized legislative formulas were immunized from all constitutional challenges. But it said little on how any fee limitations should be determined. Both of these issues were addressed in part by California’s Mitigation Fee Act (§ 66000 et seq.), whose purpose was to ensure that these exactions or impact fees were not diverted to cover the common expenses of the community. The intermediate California court held that the legislative fees were not subject to constitutional scrutiny and thus it never addressed the second question.

That court’s answer to the first question was wrong. There is simply no reason why the legislature formula should be immune to review. The takings clause of the United States Constitution applies to all government entities, state and federal, whether they delegated this authority to a planning board or choose to exercise it by itself. In both cases the takings clause’s function, as held in Armstrong v. United States (1960), is as follows:

The Fifth Amendment’s guarantee that private property shall not be taken for a public use without just compensation was designed to bar Government from forcing some people alone to bear public burdens which, in all fairness and justice, should be borne by the public as a whole.

Newcomers and owners of undeveloped land normally do not vote in local elections, and thus are an obvious target for local politics, just as are out-of-state landlords. They receive little or no protection from either state legislatures or local planning boards, so the best solution in all cases is to apply the same standards to the county’s action whether it sets the fees itself or delegates it to a subordinate body.

But do these impact fees make sense? The calculations just do not add up. The problem lies not in using a standard formula but in using the wrong standard formula. Why tie the mitigation fees to the issuance of housing permits in the first place? Houses as such do not use the roads; vehicles do. Tax vehicles instead (and some people own none and others own three or more). And then ask why these unspecified capital costs should be borne by new arrivals. Wholly apart from these exaction fees, the county imposes other taxes on a pay-as-you-go basis to maintain and operate the roads, so why force the newcomers to pay a special fee, given that they also pay their pro rata share of expenses, just like other residents? Presumably, the county could identify some special burden that newcomers impose. But why $23,000? At the very least, the total budget needs a hard look to quantify the extra fees. If there were 100 new building permits charged under this regime, the sum would be a hefty $2,300,000—an undue amount, to say the least.

The weary skeptic might ask, why elevate this cost allocation matter to a constitutional level? To see why, look at some of the resource misallocations that arise if the problem is either ignored or gutted. In Nollan, referenced above, current landowners were told that they could receive the building permit to tear down their beachfront shack to make way for a real home only if they deeded at no cost to the Coastal Commission a lateral easement across the front of their lands to let ordinary citizens walk from one public beach to another on the opposite side of their land.

Assume that the easement reduced the Nollans’ property value by $50,000 just when the new permit increased the property value by ten times that amount, or $500,000. Why not allow, as Justice Brennan had urged in Nollan, this apparent win/win transaction?

To see the flaw in the argument, note that the social objective of any government  transaction is to produce social gains—which is exactly what Brennan’s conventional argument misses. The correct comparison should be whether the value of the easement to the public is increased by an amount greater than the loss imposed on the private owner.  That question is not answered by observing that the permit is worth more to the landowner than the denial of the lateral easement. He has to take the deal. So, now suppose that the easement is worth only $20,000 to the public, then the transaction should not go through, given the Nollans’ loss of $50,000. Conversely, if the easement is worth $80,000 to the public, then the Coastal Commission can afford to compensate the property owners to eliminate their loss. Nollan therefore involves the abuse of the monopoly power and not the “out and out” coercion that Justice Antonin Scalia thought compromised the power. So, the so-called unconstitutional-conditions doctrine—the government cannot demand a surrender of a constitutional right in exchange for a private benefit—is needed to prevent the state abuse of its monopoly permitting power.

In  Dolan, the City of Tigard (Oregon) would have created similar social losses if it could impose heavy conditions on Florence Dolan when she wanted to expand the asphalt footprint of her store. The supposed justifications were two: first, that the increase in impervious surface would increase water runoff, for which it is proper for the city to make Dolan create an offsetting greenway on her property. Second, Dolan had to take measures to offset the increased traffic to her store by constructing a bike and walkway on her property. Neither of these conditions look wholly unrelated to her land use project, but why should she bear the entire cost? Dolan did not create the increased runoff, which came from other sources. So why do they or the city escape any cost for the mitigation? Similarly, the proposed walkway and bike path benefit all the people who went past the store but not to it. The failure to apportion the costs could block or delay the creation of a socially beneficial project, which is why Chief Justice Rehnquist was correct in finding that the nexus and proportionality tests were not satisfied.

The third case is Koontz v. St. Johns River Water Management District (2013), where the district sought to condition issuing a building permit on wetlands on the willingness of Coy Koontz to both surrender a large fraction of his land to the district and accept a conservation easement on the rest. In addition, the district insisted that Koontz hire contractors to improve district-owned wetlands several miles away. These conditions are suspect on matters of both ends and means. First, the repair of distant district property should be paid for out of general revenues, since Koontz neither got any special benefit from nor imposed special costs on the district that justified imposing that cost on him. The control of discharges from his own project presents a closer issue on the relations of means to ends. Clearly, the prevention of pollution is a legitimate government end. But the means chosen, as in any nuisance case, has to be proportionate to the peril. So, suppose in this case a simple settling pond could remove 99 percent of the pollutants. It would impose an undue burden to require a tenfold increase in expenditure to remove the last 1 percent when a modest effluent tax could cover the loss.

These three cases teach us that it is possible to articulate a coherent system that does not penalize new entrants or new projects. The hard question, given the meandering oral argument before the court, is whether the Supreme Court will write a principled opinion to rein in the abusive practices that state and local governments all too often impose on new developments.

© 2023 by the Board of Trustees of Leland Stanford Junior University