project that the taking enabled. However, this deduction occurs only to the extent that the remaining portion receives an idiosyncratic, ‘special’ benefit from the project. Merely receiving the same ‘general’ benefit as everyone else in the community receives from the project does not produce an offsetting reduction in compensation for the taken portion.63

Again, the intuition behind this doctrine seems straightforward: if the landowner receives a special benefit from the project, then that benefit diminishes the extent to which the taking has left the landowner worse off than everyone else in the community. However, as just noted, generally shared benefits do not have that effect. Hence, sharing in a general benefit does not diminish the amount of monetary compensation owed.

So the law governing partial physical takings is clear and consistent. But note that it is common in American law today to think of property ownership as a ‘bundle of rights’, and of property regulations as merely removing one element, or a few elements, from the bundle.64 On that view, a regulation seems to have an effect quite similar to a partial taking, which also removes a portion of the entitlements which a property owner had formerly enjoyed. That structural similarity might naturally lead us to expect that the rules for compensation would be the same in both cases—monetary compensation would be decreased only for special benefits that the owner received from the government action, not for general advantages received. Yet once again the opposite is true when general reciprocity of advantage is invoked to justify denying monetary compensation to burdened landowners. Such a stark inconsistency begs for a justification, but no obvious candidate is available.

5. Reciprocity and Respect

Since ‘general’ reciprocity does not provide a coherent justification for the permissibility of imposing restrictions on property owners without paying monetary compensation, and the requirements for ‘specific’ average reciprocity of advantage are likely to be satisfied only rarely, appeals to average reciprocity of advantage seem, at least at first glance, unable to justify the sorts of property restrictions for which they are invoked.

The question then is what we should conclude from this inability. There are three possibilities. We might conclude that these property restrictions must be illegitimate, since they do not provide the required average reciprocity of advantage. Alternatively, we might conclude that we should reject the presence or absence of average reciprocity of advantage as a test for the permissibility of property

63See Sackman et al. 2012, vol. 8A, s. G16.04[1].

64An influential minority of commentators has challenged this assumption in recent years. See e.g. Merrill and Smith 2000 and Penner 1997. However, they remain a distinct minority, and for our purposes it is not necessary to decide whether the ‘bundle of rights’ conception is in fact the best way to understand the nature of property. We need merely note that the conception is widely held.

restrictions. According to this view, outside of regulations which address coordination problems, attempts to justify governmental imposition of burdens on property owners by appealing to average reciprocity of advantage (and related concepts) are simply mistaken; the concept cannot do the justificatory work that is asked of it. Finally, we might instead conclude that the concept of average reciprocity of advantage has been misunderstood and needs to be reconceived, but that a properly understood version of the concept in fact can justify those governmental impositions. Let us examine each of those possibilities in turn.

5.1 Accepting the conclusion

One possible response to the recognition that average reciprocity of advantage will not reliably compensate owners of property burdened by government restrictions is to conclude that those restrictions are therefore illegitimate unless the government pays monetary compensation to make the burdened property owners whole. This is the most straightforward conclusion to draw from average reciprocity’s problems, and one which might appeal to people who are generally averse to government restrictions on property.

However, it faces an important difficulty: the number of restrictions that are likely to prove impermissible by this standard is so large—and so much larger than existing legal practices currently assume—that embracing this conclusion would call into question the plausibility of the entire approach which had led to such sweepingly revisionist consequences. The relevance of average reciprocity of advantage to questions about the legitimacy of government actions is not a logically necessary truth, like the truths of arithmetic. It rests on certain assumptions, and, as with any argument, if the conclusions that follow from these assumptions seem too implausible, the proper lesson may well be that we should reject some or all of those assumptions rather than that we should accept a host of counter-intuitive conclusions.65

Of course, someone who has independent grounds for welcoming those conclusions—someone who finds those conclusions completely plausible—would have good reason to embrace the assumptions which lead to them. In this case, some libertarians of a particularly austere sort might fall into that category. But even this sort of person would then have to provide some explanation of why so many

65 Avoiding this very problem is in fact the role that appeals to implicit in-kind compensation play in Richard Epstein’s theory of the property rights. Epstein begins with strong assumptions about the rights of property owners, including a requirement that regulations must be accompanied by compensation which makes regulated parties whole except when the regulation either prohibits a nuisance (which Epstein further limits to interferences involving physical invasions) or is justified ‘by the doctrines of consent and assumption of risk’. Epstein 1985, 198. This set of assumptions, taken alone, would entail that a vast array of common government regulations, such as zoning, would be constitutionally impermissible. Therefore, Epstein’s theory needs some mechanism for explaining how regulations which do not fall within his exceptions but which are widely accepted as legitimate can be justified within his theory. The notion of ‘implicit in-kind compensation’ fills this role. Recognizing the limited frequency of actual reciprocity of advantage therefore poses a significant challenge to Epstein’s argumentative strategy.

other people over extended periods of time have been mistaken in their assessment of those conclusions’ plausibility. Meanwhile, everyone who lacks an independent commitment to those conclusions has good reason to compare the relative plausibility of the conclusions to the relative plausibility of the assumptions and to reject whichever is less plausible. In this case, long-standing public and judicial beliefs about the permissible scope of government regulation and the legitimacy of eminent domain are likely to prevail over relatively novel and arcane theoretical analyses, if we are asked to choose between them.

5.2 Rejecting the premisses

This recognition naturally leads us to the second possible option, rejecting one or more of the theoretical assumptions which led to analysing the permissibility of property restrictions in terms of average reciprocity of advantage. Two main assumptions are the likely candidates here. One is the assumption that if a restriction on property is justified, then unless the restriction fell within the ‘police power’ the owner who is burdened by that restriction must have received full compensation for the costs of that burden, i.e. must have been ‘made whole’. The second assumption is that average reciprocity of advantage has some role to play in justifying property restrictions at all.

Rejecting this second assumption might initially seem to be the most natural response. After all, if a concept (such as average reciprocity of advantage) has been introduced to solve a particular problem (such as the assumed need for payment of full compensation to justify government-imposed burdens on property owners) and the concept fails to solve that problem (such as by failing to provide reliable compensation), then a straightforward reaction simply is to discard the inadequate solution.

However, we should not be too quick to choose this option, for two reasons. The first is that if we do reject the relevance of average reciprocity of advantage, and thus retain the assumption that non-police-power restrictions on property are permissible only to the extent that burdened property owners are made whole, then we still face the basic problem of justifying the imposition of regulations unaccompanied by monetary compensation, since there is no obvious alternative source of compensation to make owners whole. Hence, rejecting the second assumption would once again require denying the legitimacy of a vast range of regulations that are widely accepted as legitimate, and as we noted above, that conclusion comes at a heavy cost in plausibility.

Moreover, there is a second reason to hesitate before consigning average reciprocity of advantage to the dustbin of irrelevance: as we saw earlier, both courts and commentators have found the concept sufficiently compelling that they have often invoked it in this context. Thus, it is reasonable to suspect that there must be something to the idea, even if it has typically been misunderstood.

If so, then the more plausible approach is to reject the assumption that burdened property owners necessarily must be ‘made whole’ if restrictions on their property are to be permissible. (Again, assuming that exercise of the ‘police power’ is not involved.) This approach would assert that the real mistake of standard average

takings context, however, does not easily fit that paradigm, since the burdened owner will often have done nothing wrong. That certainly is true in the case of ordinary physical takings or zoning, and may well be true in the case of other regulations. (Regulations which are merely the exercise of some ‘police power’ will straightforwardly fit the paradigm, but, as we have seen, justifications of those regulations typically do not make appeal to average reciprocity of advantage in the first place.) The only ‘wrong’ that these owners have done is to have preferences or risk tolerances that differ from many other property owners’, and that is no wrong at all.

The question then is why merely partial subjective compensation may sometimes be appropriate, even when the burdened owner has done nothing wrong.67 Fully developing a general account of how the interaction of chance with the web of civic duties within which the institution of property sits may justly result in the government’s imposing burdens upon landowners without also providing a subjectively equal measure of compensating benefits is a large task necessarily beyond the scope of this chapter. However, it is possible here for us to see how the presence of reciprocal advantages can contribute to such a general account.

5.4 Reciprocity and shared sacrifice

To see how the presence of reciprocal advantage, even subjectively partial reciprocal advantage, can help justify imposing restrictions on property owners, we must broaden the focus of our discussion to include not only the owner who is burdened by some restriction but also the other members of society who benefit from that restriction. Consider a touchstone principle of American constitutional law governing compensation for physical takings, described in the oft-quoted words of Justice Hugo Black’s opinion in Armstrong v United States: ‘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.’68 Note that the principle’s emphasis is not on leaving those burdened by takings at least as well off as they were before the taking—the principle says nothing about making them whole. Instead, the court focused on ensuring that burdens are shared, that is, on ensuring that the community participates in bearing the burdens of public life.

67 I have argued elsewhere (Lee 2013) that governments that take property through eminent domain may sometimes justly pay merely partial compensation for the subjective losses suffered by owners of the taken property, even when considerations of practicality or administrability do not require paying less than full compensation.

68 Armstrong v United States 1960, 49. William Treanor has traced the considerable influence of the ‘Armstrong principle’, noting that it has ‘received a remarkable degree of assent across the spectrum of opinion’. Treanor 1997, 1153; see also at 1153 nn. 17–22 and accompanying text. The basic idea was not original with the Armstrong court. For one historical antecedent, see Vanhorne’s Lessee v Dorrance 1795, 310 (‘[N]o one can be called upon to surrender or sacrifice his whole property, real and personal, for the good of the community, without receiving a recompence in value. This would be laying a burden upon an individual, which ought to be sustained by society at large’).