66 Am. U. L. Rev. 1433 (2017)
*Associate Dean, and Professor of Law, Vanderbilt Law School.
This is a Response to Bethany R. Berger’s recent Article, The Illusion of Fiscal Illusion in Regulatory Takings. In that Article, Professor Berger argues against the view that governments should be forced to compensate for regulatory burdens because they suffer from fiscal illusion and will only internalize the costs that they, in fact, have to pay. She demonstrates that property taxes already provide a mechanism through which governments internalize both the costs and benefits of their property regulations, and that compensation for regulatory takings is therefore unnecessary and even perverse for creating efficient regulatory incentives. This Response argues that she is correct and that her criticism joins many others before it. However, despite these significant criticisms, fiscal illusion continues to inform takings theory. This Response ultimately demonstrates that even if Professor Berger is correct, her proposed responses do not entirely address the problem that compensation will over-deter government regulations, and so this Response proposes both municipal insurance and liability for regulatory inaction as potential interventions.
In her recent Article, The Illusion of Fiscal Illusion in Regulatory Takings, Professor Bethany R. Berger takes aim at fiscal illusion as a justification for regulatory takings. It should be an easy target; people have shot at it before and hit it repeatedly. And yet it somehow does not die. The concept of fiscal illusion is based on an alluring intuition that governments should internalize the costs of their actions, and that failing to require compensation for regulatory burdens will lead to over-regulation. As many critics have pointed out, however, the intuition does not seem to withstand close scrutiny. Governments do not internalize costs the way private actors do, and the fit between compensation and regulatory incentives is so imprecise as to be meaningless.
Professor Berger adds a new criticism, attacking fiscal illusion on its own terms. She points out that property taxes are, in essence, a mechanism by which local governments internalize both costs and benefits and that a requirement to pay compensation will distort any calibration between the two, putting a heavy thumb on the scale against government action. This is a good point and builds nicely upon claims in the existing literature. But fiscal illusion is an undead theory, and in the face of its persistence, it is important to consider other ways in which the legal system can respond to the over-deterrence of compensation. These include a more robust form of municipal insurance or, absent that, an offsetting compensation requirement for regulatory inaction that balances out the threat of traditional takings liability.
Part I of this Response canvasses the critiques of fiscal illusion. Professor Berger does some of this work herself, but it is useful to see the full arsenal arrayed against this zombie theory. Part I is divided between external and internal critiques, and Professor Berger’s falls into the latter category. Part II looks at potential responses to the problem of over-deterring government action. Building on my own prior work, it proposes both takings for regulatory inaction and insurance as ways of addressing the problem of under-regulation and under-enforcement of existing regulations.
I. The Assault on Fiscal Illusion
According to Professor Berger’s succinct summary, the use of the term “fiscal illusion” in takings literature dates back to a seminal article by Lawrence Blume and Daniel Rubinfeld, although the concept is older. The phrase is invoked for the idea that “the government does not take into account the societal costs of governmental actions unless the government itself pays for them.” It therefore reflects the familiar economic intuition that the law should force people—or organizations—to internalize the costs of their actions in order to induce efficient decision making. Externalized harms will tend to encourage too much of an activity relative to its overall impact, whereas externalized benefits will result in too little of an activity. When it comes to private law—property and nuisance law, for example—this idea is intuitive and powerful. Applied to the Takings Clause, it suggests that the government should bear the costs of its actions and that forcing the government to pay compensation for the regulatory harms it imposes will induce efficient regulatory incentives. The fiscal illusion theory has had enormous influence on takings theory. However, it is not at all clear that forcing the government to pay compensation will, in fact, force it to bear the costs of its actions in any meaningful sense. Moreover, the theory of fiscal illusion misses important additional dynamics affecting both regulatory and investment incentives. Broadly, these can be grouped into external critiques that challenge the assumptions and implications of fiscal illusion and internal critiques that challenge fiscal illusion on its own terms. These critiques are introduced below.
A. External Critiques
External critiques of fiscal illusion challenge the assumptions and implications of the theory and include, for example, invoking public choice theory, focusing on the heterogeneous sources of revenue for paying compensation, and problematizing the resulting investment incentives.
- Public choice
Key to the traditional economic account of externalities is the assumption that people—and corporations—are rational economic actors. Forcing private actors to bear the costs of their conduct will lead to more efficient outcomes precisely because they are price sensitive. The same is not necessarily true of governments. Numerous people have argued that government actors are primarily motivated not by the goal of maximizing wealth but instead by the goal of maximizing political power. According to public choice theorists, government officials act in their own self-interest and are primarily motivated by re-election as opposed to public expenditures. Sometimes the two are aligned, as when officials get political credit for controlling costs, but often they are not.
In some larger units of government—such as the federal government, states, cities and larger municipalities—political pressure is often wielded by special interest groups more than by voters or taxpayers as a whole. Small motivated groups, facing relatively high per capita stakes, enjoy organizational advantages over larger and more diffuse groups. This is important because compensation for regulatory takings is paid out of taxpayer-funded public coffers. Instead of forcing governments to internalize the costs of their actions, paying compensation for burdensome regulations can spread costs over a broad tax base, while buying the acquiescence of the one group well situated to raise genuine political objections: the burdened property owners. However, it is not clear that a compensation requirement will generate any meaningful political opposition to burdensome regulations when taxpayers are not especially sensitive to small changes in their tax bills or when the cost of this kind of compensation is hidden in the overall budget. There may be mechanisms by which regulatory takings generate political costs, but they are much more attenuated and less direct than the standard fiscal illusion account would suggest.
I have previously attempted to identify one such mechanism. In Big Differences for Small Governments: Local Governments and the Takings Clause, I argued—contrary to the public choice intuition—that compensation may, in fact, force small local governments to internalize the costs of their actions because both the costs and benefits of local regulations are borne by local homeowners who vote. The benefits of regulations and government projects are reflected in local property values, while the costs of compensation are paid largely through property taxes and so are capitalized into property values. In other words, local homeowners internalize both the costs and benefits of government actions.
According to William Fischel, homeowners dominate the politics of small local governments; they are politically active, benefit from social networks, and are highly attentive to local property values. As a result, local officials are closely attuned to these “homevoters,” who in turn are motivated primarily by preserving their property values. Put this all together, and the traditional cost internalization story of compensation for regulatory takings actually works surprisingly well in small local governments. Costs and benefits are both captured in property values, which become a kind of political yardstick for local government success.
This argument, however, was not intended to repudiate the public choice criticism of just compensation—quite the opposite. In defending the traditional economic account of compensation’s effects for small local governments of a particular character, this argument reinforced the public choice critique elsewhere. Identifying the mechanism for feedback between compensation and political pressure in small local governments reveals the absence of such a mechanism in larger governments—or in governments not dominated by Professor Fischel’s homevoters. Moreover, I argued that even in homevoter jurisdictions, local officials may actually be risk averse, and so the effect of compensation may be to deter government actions that would, in fact, have been net beneficial. While I return to the problem of risk aversion in greater detail below, the more fundamental insight here is that public choice theory challenges proponents of fiscal illusion to develop a mechanism by which economic costs are translated into political costs. It may be possible to develop such a theory in specific regulatory contexts, but overall, the public choice criticism of fiscal illusion is quite trenchant.
2. The problematic sources of revenue
The political ramifications of compensation also depend upon how a government raises revenue. The idea of fiscal illusion implicitly assumes that the government is a kind of unitary economic actor. Facing the prospect of compensation will force the government to make trade-offs, assess opportunity costs, and make more rational decisions about regulatory priorities. But the effect—and, therefore, the politics—of compensation will be quite different depending on the sources of revenue.
This is most starkly on display in Professor William Fischel’s granular account of Detroit’s decision to condemn property in the famous case of Poletown Neighborhood Council v. City of Detroit. While Poletown involved eminent domain and not regulatory takings, the dynamics surrounding revenue are the same. According to his analysis, one of the central features of the Poletown case was the federal government paying the lion’s share of Detroit’s land assembly. As a result, in Professor Fischel’s colorful language, “opposing the Poletown project looked like shooting Santa Claus.” Generalizing from this specific example: the source of the funding for compensation undoubtedly affects the political economy of local decision making.
These potentially distorting effects do not depend on federal or state earmarking either. The dynamics surrounding revenue sources can be important more broadly. Consider a local government that levies an uneven property tax. New York City, for example, treats single-family homes much more favorably than apartments for tax purposes. The result is that owners of multi-million dollar brownstones may pay far less in taxes than owners of equivalently valuable apartments. Paying compensation from property tax revenue in New York City means that taxpayers unevenly bear the actual burdens of compensation. Add other sources of revenue, like sales or income tax, and the connection between the award of compensation and the sources of funding becomes increasingly tenuous. The sources of revenue mean that the costs of compensation are, in fact, spread unevenly throughout a jurisdiction.
Some might object that money is fundamentally fungible and the source of revenue does not matter for the underlying claim of fiscal illusion. The government will pay out from a limited pile of money, and that is enough. The problem runs deeper, however, because of the mismatch between regulatory costs and benefits. As sources of municipal revenue diversify, there is an ever-increasing divergence between the benefits of a local regulation and the source of revenue to pay compensation. Zoning changes, for example, will easily translate into property values, and so property taxes are at least a reasonable nexus for balancing costs and benefits (even if both costs and benefits are unevenly distributed—a point taken up again below). But, if neighbors enjoy the benefits of a downzoning while the costs are borne primarily by businesses (paying sales tax), out-of-town visitors (paying a hotel tax), or others, then compensation does not create even a rough balancing of costs and benefits. It simply pits two groups against each other staking out clear winners and losers. And this is even before problematizing how local governments internalize benefits.
3. Investment incentives
Another criticism of fiscal illusion is the effect of compensation on property owners’ investment incentives. As Professors James Krier and Michael Heller recognized in an insightful treatment of the problem, the compensation rule that generates efficient regulatory incentives may not be the same rule that generates efficient investment incentives. Whether compensation is important to force the government to internalize the costs of its actions, the promise of compensation may have perverse effects on property owners by inducing a form of moral hazard.
In its simplest form, the moral hazard problem in regulatory takings is that compensation can be viewed as a kind of mandatory insurance against the most significant costs of regulatory change. If that insurance were to become more widely available—as the logic of fiscal illusion compels—then it would allow property owners to ignore risks that they should, in fact, consider. Property owners should not ignore foreseeable risks of regulatory changes when making their investment decisions. For example, imagine owners of property near an airport deciding whether to develop condominiums or a shopping mall. Imagine that the condos are the more valuable use but that there is a chance the FAA may re-route airplanes directly over the property. The property would then be much less valuable for residential than commercial uses. Discounting the expected value of the various choices by the risk of airplane overflights, the shopping mall might become the more valuable use. The promise of compensation for such a regulatory change would allow the property owners to ignore those risks, however. Even though the condos are suddenly the less valuable choice because of the possibility of increased airplane noise, the developer—and, to some extent, even purchasers of the units—may be indifferent so long as they receive compensation for this diminished utility. Better, of course, would be for the developer to take this risk into account. Ignoring the risk will create deadweight costs and also increase the costs of compensation to the government.
This problem led Professors Krier and Heller to propose an innovative solution: disaggregating the amount the government has to pay from the amount that property owners receive. Their approach would effectively allow compensation to be calibrated to address fiscal illusion on the one hand, and to address moral hazard on the other. While theirs is a creative approach, it is most certainly not the law. It also reveals—at least implicitly—that addressing fiscal illusion can create different problems for investment incentives, unless there are radical changes to the compensation regime.
B. Internal Critiques
In addition to these external critiques, scholars also challenge the internal mechanisms of the fiscal illusion theory. These focus on the problem of lumpy costs, the lack of symmetry between costs and benefits, and temporal problems.
- Lumpy costs
Even taking the fiscal illusion story seriously on its own terms, Professor Daryl Levinson demonstrated that the heterogeneous distribution of costs and benefits can produce inefficient outcomes. This is true even under the hyper-stylized conditions that typically drive examples of fiscal illusion.
Scholars often explain fiscal illusion by imagining a stylized municipality with an arithmetically simple number of people, say 100 property owners choosing whether to downzone (or otherwise burden) one individual property owner. Readers are asked to imagine that the regulation will impose $1,000 (or some arbitrary number) worth of harm and conclude that if the regulation is worth at least $10 per property owner, then it is efficient. In this story, compensation is said to put the costs and benefits to the test. If property owners must pay the $1,000 of regulatory harm, they will only support the regulation if it is worth at least $10 per owner.
Professor Levinson, however, demonstrated the problem with this analysis by adding an additional but realistic wrinkle: that the benefits are spread unevenly among the property owners. In a majoritarian political system, using the same example as above, if fifty-one of the 100 property owners value the regulation at $11, and forty-nine of them value the regulation at only $1, then the government will still adopt the regulation even though it generates only $610 of benefits and $1,000 worth of costs. Those who value the regulation more can spread the costs over the entire tax base. Therefore, even a compensation requirement in a perfectly representative government will not necessarily produce efficient results if the benefits are too lumpy.
Notice that this calculation relies on a slightly different and more sophisticated version of the fiscal illusion claim. Instead of viewing the government as some autonomous actor with its own set of preferences, it assumes a representative government. Here, determining a government’s regulatory incentives requires understanding the priorities of a majority of voters. The government’s motives are then a reflection of the polity’s preferences, which includes the possibility of interest groups and heterogeneous voter preferences.
This is different from public choice theory, of course, which focuses on the preferences and incentives of individual government officials. Public choice theory therefore accounts for minoritarian rule, whereby local officials do not follow majority preferences but instead the preferences of well-organized special interest groups. Levinson’s account, in contrast, remains majoritarian, but more realistically so by recognizing that different people and different groups may value government actions differently. This does not apply only to the cost side of the ledger, however. Adding an explicit focus on regulatory benefits introduces an additional critique.
2. The problem of symmetry
Professor Levinson—in the same article discussed above—offers a new gloss on the familiar public choice critique. Fiscal illusion assumes that governments only internalize costs that have a direct budget impact, and yet, seems to assume that governments internalize benefits politically or through some other opaque mechanism. In short, fiscal illusion presumes a curious asymmetry in how governments internalize costs and benefits. If a government internalizes benefits without being able to capture them in its balance sheet, then the same should be true of costs as well. Traditional fiscal illusion analysis focuses only on regulatory costs.
The implicit answer in traditional fiscal illusion analysis is that governments internalize benefits politically. The motivation for a regulation comes from the benefits it will produce—benefits that apparently flow to local property owners and groups with political power. But this devolves quickly into the familiar public choice critique described above because, if political pressure motivates governments to generate regulatory benefits, the same forces should motivate governments to avoid regulatory harms. While the politics of harms and benefits may be different in any particular setting—more diffuse or more concentrated, or even falling on groups with more or less political force—there is no reason in the abstract to think that governments internalize harms and benefits in fundamentally different ways. In other words, if governments internalize benefits without any immediate impact on the budget, then the same should be true of costs as well. Compensation is therefore unnecessary, at least for the purpose of inducing efficient regulatory incentives.
The argument for symmetry can cut in the other direction as well. If the government internalizes costs through its budget, then it must also be allowed to recover regulatory benefits for precisely the same reason. But what would it mean for governments to capture the benefits of their regulations? In the strongest form, it could look like the radical proposal for “givings,” allowing the government to charge the beneficiaries for regulatory largess. That, of course, is not the law. Alternatively, and more plausibly, some scholars have recognized that property taxes provide a mechanism by which local governments can capture at least some of a regulation’s benefits. If those benefits are capitalized into property values, then they will produce greater property tax revenue. From the government’s perspective, costs are paid through compensation and benefits are recaptured over time through increased property taxes. But a moment’s reflection reveals that this is still not symmetrical; property tax revenues may indeed allow a government to capture some regulatory benefits but not their full worth because property taxes are levied as a small percentage of property values.
3. Berger and the temporal problem of taxes
This final observation leads nicely into Professor Berger’s contribution, which makes a new version of the symmetry argument. Professor Berger argues that property taxes are the mechanism by which local governments internalize both benefits and costs. Regulations that increase local property values will increase tax revenue, and regulations burdening property will have the opposite effect. Importantly, however, this mechanism for cost internalization occurs even without a compensation requirement. Clearly, governments internalize costs and benefits through tax revenue to a different extent than the property owners experience them. The translation from property values to property taxes occurs through the tax rate and is mediated by imperfect assessments. But fiscal illusion is concerned only with regulatory incentives, and so long as costs and benefits are internalized to the same extent, it does not matter whether both are fully or only partially borne by the government.
Professor Berger, therefore, argues that compensation is not only unnecessary to induce efficient regulatory incentives, but that it will have the perverse effect of heavily weighting the cost side of the ledger. Governments imposing regulatory burdens will not only bear the cost through lower property tax revenue, they will also have to pay a lump-sum in compensation that must be recouped slowly through the increased property taxes from benefitted property. Instead of inducing efficient regulatory incentives, compensation in this view is a heavy thumb on the scale against even net beneficial regulations.
Professor Berger’s argument really contains two distinct but intertwined attacks on fiscal illusion. The first is simply to observe the symmetry of costs and benefits reflected in property taxes, and therefore the “additional” cost that compensation represents. The second is the problem of an inter-temporal imbalance, whereby costs of compensation are borne all at once while benefits through property taxes are achieved only slowly, over time. It is important to keep these conceptually separate.
The symmetry argument is elegant and serves as an effective riposte to the conventional articulation of fiscal illusion that views the government as its own independent entity, exclusively concerned with its own budget. However, it is not entirely responsive to the kind of more nuanced articulation offered by Professor Levinson, for example, who views the government as the conduit for expressing voter preferences. It is not the government that must internalize costs and benefits, but rather the voters who fund the government. Forcing the government to pay is merely the mechanism for requiring regulatory winners to bear the costs—through their taxes—of paying off the losers in the regulatory process.
Here, Professor Berger’s argument for symmetry becomes less conclusive. In this kind of majoritarian polity, property taxes do not necessarily counterbalance in the way she suggests. For example, in a stylized municipality, imagine a measure that will increase the property value of 100 property owners by $100 each (for a total benefit of $10,000). However, it does so at the expense of a single burdened property owner, who will see a reduction in her property value of $50,000. If the property tax rate is 5%, then property taxes will simultaneously increase by $500 per year (to reflect the increase in property value of the 100 benefitted properties) and decrease by $2,500 per year (to reflect the diminution in value of the burdened property), for a net loss of $2,000 per year.
So far so good. Focusing on the government as its own entity, with its own incentives, the government will not enact this measure even in the absence of a compensation requirement, just as Professor Berger suggests. But, focusing on votes instead of the budget makes the story more complicated. Now, 100 taxpayers stand to benefit by $100, and so would presumably vote in favor of the measure, even though the costs to the individual burdened by the measure are far greater than the aggregate benefits. Even in a perfectly representative government—putting aside public choice theory—voters in this example will still approve an inefficient government regulation that benefits more people than it burdens.
However, maybe the analysis should not stop there. Focusing on the taxpayers instead of on the government means taking into account that taxpayers will also have to make up the foregone tax revenue resulting from the reduction in value of the regulated property. In this example, the 100 benefitting property owners will presumably have to cover the $2,000 reduction in annual property taxes to the tune of an extra $20 per person. Is the $100 increase in underlying property values worth another $20 per year in taxes? Depending on discount rates, probably not. But this is the muddier analysis that needs to be navigated as the focus shifts from the government fisc for its own sake to the taxpayers footing the bill.
Add a more heterogeneous tax base—as Levinson does—and it becomes murkier still. Now imagine that the 100 benefited taxpayers are a majority, but nevertheless a subset of 150 taxpayers in town. And imagine further that they own only 25% of the property tax base. In other words, 75% of all local property by value is owned by the remaining fifty people, and they will not benefit from the measure at all. Property tax revenue will still decrease by net $2,000 per year, but the 100 benefited property owners will only have to contribute $500 to cover the shortfall. The benefited property owners will have to pay an additional $5 apiece in taxes—a price they may happily pay for the $100 benefit. The wealthier taxpayers will have to contribute the extra $1,500 for a measure they do not value. And yet, in a majoritarian system, the measure may now pass.
It is easy to play with the numbers and produce all sorts of different results. Nevertheless, the overall point is easy enough to see: property taxes will not necessarily do the work Professor Berger suggests once costs and benefits are unevenly distributed, even in a perfectly representative political system. As Levinson points out, this concern exists with regards to a more traditional compensation analysis—one not mediated through property tax rates. Professor Berger’s account fares no worse, but it does reveal that her argument for the symmetry of costs and benefits in the property tax system falters when tax burdens and regulatory benefits are unevenly distributed.
While this focus on the taxpayer instead of the government undercuts some of Professor Berger’s symmetry argument, it actually reinforces her other concern that the payment of compensation creates a lump-sum cost, whereas regulatory benefits are spread out slowly over time. The inter-temporal mismatch between costs and benefits can create a problem for taxpayers who may face a sudden spike in their property tax bills or for government officials faced with immediate cash flow pressures. Indeed, this is a profound problem for regulatory incentives.
II. Addressing the Temporal Mismatch
It should not be. If the concern is spikey costs and inter-temporal mismatches between costs and benefits, there is a standard recipe for responses that includes both debt and insurance. As it turns out, neither debt nor insurance is an effective response, but the reasons are not obvious and are worth considering in some detail.
Regulatory takings liability is hardly unique in imposing large one-time costs that can only be repaid over time. Indeed, this bears a surprising similarity to infrastructure spending, say on a new school, library, or bridge. These sorts of investments also create problems of inter-temporal mismatch where the government must incur all of the up-front costs, but benefits accrue far into the future. This is precisely the same dynamic that Professor Berger identified that prevents efficient or net-beneficial regulations. Those regulations create immediate costs but long-term benefits through improved regulations. When it comes to infrastructure, there is a straightforward solution to realigning those costs and benefits over time: debt financing. By paying for a school expansion or other capital investment with municipal bonds, the government can—in effect—spread the costs of the investment over time. Taxpayers in the future, who will continue to benefit from the investment, will also have to pay their share by continuing to service the debt. This is the normative justification for financing infrastructure with debt in order to align costs and benefits over time.
There is no reason in the abstract why the same should not be possible for regulatory takings liability. Faced with a large up-front cost that will generate long-term regulatory benefits, the government could float a bond to cover the costs of compensation. The problem of spikey costs would disappear, and regulatory incentives would therefore become realigned. The possibility of financing regulatory takings compensation should largely undermine Professor Berger’s concern.
While this makes good sense conceptually, there are doctrinal and practical problems with this approach. For one, almost every state limits municipal indebtedness, usually to some fraction of overall assessed property value in the jurisdiction. The reason is that municipal debt raises a particular kind of inter-temporal problem by allowing local officials to raise money today while pushing the burdens into the future. This is appropriate when the investment creates long-term benefits, but it makes much less sense where the money is used for more speculative ventures or simply to cover short-term operating costs. In those instances, public choice theory predicts that local officials will be insufficiently attentive to future concerns; they will reap the political benefits of having more money today, while forcing future taxpayers to bear most of the costs. State law limiting municipal debt should be seen as a response to this problem.
Municipal debt limits, however, pose a problem for financing compensation awards. Additional borrowing may simply not be available for some governments. And even where it is, incurring debt to pay compensation for regulatory takings claims still imposes significant opportunity costs because less borrowing authority will be available for other purposes. In short, then, municipal debt is not a cure-all to compensation’s temporality problem.
Debt—were it to be available—is a way of smoothing costs over time ex post. It converts the need for a large one-time payment into a smoother stream of payments into the future. But there is another mechanism that serves precisely the same purpose ex ante: municipal insurance. Insurance amounts to a stream of payments ahead of time to prevent having to pay a substantial lump-sum all at once. Insurance and debt are, in effect, flip sides of this same coin.
All but the largest municipalities have insurance that provides more or less comprehensive protection against a wide range of losses. Typical policies include coverage for property loss as well as for legal liability. Liability coverage includes coverage not only for tort claims, but also for deliberate policy choices under a municipality’s Errors and Omissions (E&O) policy. Where insurance is in place, Professor Berger’s concerns about the lumpiness of compensation awards simply do not apply.
In fact, however, municipal liability insurance for regulatory takings claims does not exist for most municipalities. In a strange lacunae in insurance markets, private insurance for regulatory takings falls within a nearly ubiquitous exclusion for “inverse condemnation.” Inverse condemnation is, in effect, regulatory takings by another name, and the effect of the exclusion is to eliminate the insurer’s duty both to indemnify and defend regulatory takings claims. In previous work, I demonstrated that the exclusion was difficult to understand in light of existing insurance coverage for § 1983 claims more generally and for other types of claims against local governments. I argued that the absence of insurance will induce smaller, risk averse governments to under-regulate or under-enforce existing regulations, precisely because of the inability to smooth the costs of both compensation and litigation associated with regulatory takings claims. Count this, then, as reinforcing Professor Berger’s observation. Insurance would be a response to her concern but, in fact, it is largely unavailable.
Professor Berger is correct that there are significant problems with fiscal illusion as a normative justification for requiring compensation under the Fifth Amendment. Instead of inducing efficient investment incentives, it is more likely to over-deter local governments. As a result, Professor Berger responds by turning away from regulatory incentives as an appropriate focus for regulatory takings and calling instead for a renewed focus on fairness. This is an entirely sensible move, and one with which I am fundamentally sympathetic. But having raised the problem of exaggerated costs, it is not so easy to dismiss. This zombie is not so easily retired to its grave. Professor Berger’s argument, after all, does not challenge the idea that governments are motivated by cost. Instead, she challenges the possibility of any careful balancing by governments because they will be too motivated by the costs of compensation. While we can (and should) develop new normative theories and argue for their adoption, the Takings Clause continues to disincentivize regulatory action because the costs loom so large.
This is true despite the fact that governments rarely lose. Whether a government is ultimately liable for a takings violation, litigation costs alone can impose a significant financial burden on local governments and raise the same inter-temporal mismatch that Professor Berger identified. Moreover, because of local officials’ risk aversion, even if the probability of takings liability is low, it may nevertheless have an outsized effect on local decision making. In other words, Professor Berger’s criticisms of fiscal illusion raise problems that persist even if fiscal illusion is rejected as a justification for compensation.
Therefore, turning to fairness is no panacea. Instead, so long as regulatory takings exist as a basis for liability, governments may under-regulate and under-enforce existing regulations. There are, however, some steps that are available to address perverse regulatory incentives created by the risk of compensation.
The first, and most straightforward, is to make regulatory takings insurance available to local governments. This could be done through private insurers, municipal insurance pools, or through state indemnification. The mechanism does not particularly matter for purposes of the present discussion. What is important is that insurance can address the problem of inter-temporal mismatch that Professor Berger identified by smoothing the costs of compensation awards and overcoming local officials’ risk aversion. As a result, it ensures that local officials can more rationally calculate the expected costs and benefits of any regulation. This is not assuming anything about how local governments or local officials internalize the overall costs and benefits of a regulation, but only suggesting that local officials will be freed from the specter of adverse takings judgments when deciding whether to regulate.
A second and more unorthodox response is to introduce a countervailing pressure by awarding compensation for regulatory inaction. The problem, again, is that compensation awards will have a disproportionate effect on government decision making; they will not force governments to internalize the costs of their actions but will instead serve as a heavy thumb on the scale against regulation. The ideal way to respond is to remove that pressure, for example, by extending insurance or just eliminating regulatory takings liability. But there is another form of response: add a counterweight. If the threat of compensation will discourage government officials from regulating by magnifying the costs, that can be overcome either by magnifying the benefits as well or by imposing costs for failing to act. These responses take different forms but amount to the same idea.
Magnifying benefits could, theoretically, be accomplished through the kind of “givings” mechanism described above. Allowing governments to levy fees to capture regulatory benefits would perfectly address Professor Berger’s symmetry concern as well as the problem of inter-temporal mismatch. It would not so much solve the problem as create the equivalent problem on the opposite side of the ledger. It is also quite outlandish as an actual proposal. Although the idea has been in the literature for years, no court (or government) has taken it up.
A more plausible response, at least doctrinally, is to counterbalance the regulatory disincentive by making the government liable for failing to act. This is an argument I considered at length in prior work, and it is not as bizarre as it might seem. Creating affirmative duties on the government has been part of the progressive constitutional agenda for decades. Using property and the Takings Clause to accomplish this goal is admittedly unusual and controversial, but it does not require much of a doctrinal stretch to think that governments can—in some circumstances—violate the Takings Clause by failing to change existing regulations when property is threatened by some external change in the world. In fact, property owners regularly sue local governments for failing to rezone property.
For present purposes, the important observation is not simply that the creation of omission liability is plausible; rather, that it responds in surprising ways to Professor Berger’s symmetry argument. Her concern, again, is that the compensation requirement is out of balance with the slow recovery of regulatory benefits through increased property taxes. Imposing liability for failing to act changes that calculus, however, because avoiding those costs becomes a way of capturing regulatory benefits. After all, avoiding costs that would otherwise be due has the same effect as receiving a payment; costs avoided amount to a payment received.
Neither of these changes in the law would undermine or be inconsistent with Professor Berger’s focus on fairness. To say that takings law should center primarily on distributional consequences and ensure that disfavored individuals and groups are not singled out for regulatory burdens does not mean ignoring regulatory incentives. But these concerns do change the focus. Proponents of traditional fiscal illusion worry that failing to compensate for regulatory burdens will induce inefficient over-regulation. Here, the critiques of fiscal illusion suggest that paying compensation will result in inefficient under-regulation. To the extent that fairness considerations will ever compel regulatory takings liability, we must continue to worry that governments will under-regulate and under-enforce existing regulations. Municipal insurance and liability for passive takings help to blunt those forces.
Professor Berger deserves credit for inventing a creative new weapon to fight the theory of fiscal illusion. Her Article joins the crossbows and chainsaws of existing external and internal critiques that have been deployed to resist the argument that government must always pay. Theorists have long pointed out that political costs, not fiscal costs, are the ones that matter. The uneven distribution of costs and benefits makes any compensation regime an imperfect check on inefficient regulations. And fiscal illusion relies upon an implausible, even bizarre, assumption that governments internalize costs and benefits differently. Nevertheless, fiscal illusion keeps coming. Its thirst for the blood of local governments, in particular, appears insatiable, and so it continues to be invoked as a justification for expansive takings protection.
Professor Berger argues that property taxes already create a mechanism by which governments internalize costs and benefits. Not only is compensation unnecessary to induce efficient regulatory incentives, but it also over-deters local governments. She makes a compelling argument, but it is unlikely to stop the fiscal illusion zombie’s advance. As a result, it is important to look to other responses to the problem of over-deterrence. Municipal liability insurance and takings for regulatory inaction turn out to be surprisingly powerful responses. They are not rejoinders to fiscal illusion so much as inoculation. At the least, they may serve to restore local officials’ incentives to evaluate regulatory risk rationally. And so they may clear the way to allow for a renewed focus on other values, like fairness, that should inform the constitutional protection of private property.
. See e.g., Lawrence Blume & Daniel L. Rubinfeld, Compensation for Takings: An Economic Analysis, 72 Cal. L. Rev. 569, 573 (1984) (concluding that not all regulatory actions warrant compensation, even in the presence of “fiscal illusions”). See generally Daryl J. Levinson, Making Government Pay: Markets, Politics, and the Allocation of Constitutional Costs, 67 U. Chi. L. Rev. 345 (2000) (problematizing how governments internalize costs and benefits).
. See e.g., Berger, supra note 1, at 4 (“The efficiency argument goes like this. If governments need not compensate owners for the loss in value caused by restricting land use, they need not internalize the costs of their actions.”); Frederic Bloom & Christopher Serkin, Suing Courts, 79 U. Chi. L. Rev. 553, 576 (2012) (describing the conventional economic account of the Takings Clause); see also Richard A. Epstein, Bargaining with the State 84–86 (1993) (discussing how the just compensation requirement could be a solution to avoiding legislation that only advances factional interests); William A. Fischel & Perry Shapiro, Takings, Insurance, and Michelman: Comments on Economic Interpretations of “Just Compensation” Law, 17 J. Legal Stud. 269, 269–70 (1988) (describing the dual purpose of the compensation requirement of offering “protection to private entitlements” while “disciplining the power of the state”).
. See Levinson, supra note 2, at 377; see also, e.g., Ronit Levine-Schnur & Gideon Parchomovsky, Is the Government Fiscally Blind? An Empirical Examination of the Effect of the Compensation Requirement on Eminent-Domain Exercises, 45 J. Legal Stud. 437, 440 (2016) (“Our findings appear to call into question the prominence of the fiscal-illusion hypothesis as a single explanation for government behavior in the takings context.”); Barton H. Thompson, Jr., Conservation Options: Toward a Greater Private Role, 21 Va. Envtl. L.J. 245, 289 (2002) (reviewing evidence and arguments challenging fiscal illusion).
. See, e.g., Steven Shavell, Foundations of Economic Analysis of Law 80 (2004) (“[I]t will be socially desirable for individuals to engage less often in acts that cause detrimental external effects than is in their immediate self-interest, and to engage more often in acts that engender beneficial external benefits than is in their self-interest.”).
. Id.; see also Yun-chien Chang, Economic Value or Fair Market Value: What Form of Takings Compensation Is Efficient?, 20 Sup. Ct. Econ. Rev. 35, 87–88 (2012) (surveying theorists who have implicitly or explicitly endorsed fiscal illusion).
. See, e.g., Christopher Serkin, Big Differences for Small Governments: Local Governments and the Takings Clause, 81 N.Y.U. L. Rev. 1624, 1627 (2006) [hereinafter Serkin, Big Differences] (“[Public choice theorists] argue that paying compensation will not necessarily force the government to internalize the political costs of its actions, and it is political—not monetary—costs that matter for creating efficient regulatory incentives.”); Christopher Serkin, Passive Takings: The State’s Affirmative Duty to Protect Property, 113 Mich. L. Rev. 345, 361 (2014) [hereinafter Serkin, Passive Takings] (“Public-choice theorists have roundly and quite convincingly criticized this account’s simplest form by problematizing how governments internalize costs.”).
. See Daniel A. Farber, Public Choice and Just Compensation, 9 Const. Comment. 279, 288 (1992) (“If we adopt a public interest theory of government, internalizing a cost makes no difference.”); Lee Anne Fennell, Taking Eminent Domain Apart, 2004 Mich. St. L. Rev. 957, 995 (“While it is a convenient fiction to suppose that governments feel the pain of payments as much as private individuals do, the pain from budgetary outlays is indirect, attenuated by the operation of the political process.”); see also Levinson, supra note 2, at 377 (positing that most officials do not have to worry about the marginal effect of tax burdens); Susan Rose-Ackerman, Against Ad Hocery: A Comment on Michelman, 88 Colum. L. Rev. 1697, 1706 (1988) (“[I]f public choices are the result of the competition of various groups for political benefits, powerful groups will not need a constitutionally mandated takings doctrine in order to preserve their interests.”); Henry A. Span, Public Choice Theory and the Political Utility of the Takings Clause, 40 Idaho L. Rev. 11, 22–23 (2003) (“[T]he government is not like any other actor. It is not a simple or self-contained entity, and strictly speaking, it is both impossible and undesirable to force it to internalize either its costs or its benefits.”).
. See, e.g., Daniel A. Farber & Philip P. Frickey, Law and Public Choice: A Critical Introduction 21–24 (1991) (discussing two different model theories of how legislators get elected: maximizing appeal to constituents, or granting special interest groups a bigger role).
. See Neil K. Komesar, Law’s Limits: The Rule of Law and the Supply and Demand of Rights 61 (2001) (discussing how interest groups with small numbers but high per capita stakes have a political advantage over other groups with large numbers but low per capita stakes); see also Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups 53–57 (1965) (describing the coherence and effectiveness of small groups).
. See Farber, supra note 17, at 282 (explaining how local tax payers and property owners, in essence, purchase insurance against takings by paying higher taxes but then receive compensation after a taking).
. Serkin, Big Differences, supra note 15, at 1639 (“[M]oney for compensation will be raised from the taxpayers generally, and they constitute precisely the kind of diffuse and unorganized majority that a mobilized special interest group’s political pressure is likely to overwhelm.”).
. See id. (“[C]ompensation . . . will serve to insulate the government from the political costs that otherwise would have been at stake. With the burdened property owners out of the equation, no one else is well situated to generate significant political opposition to an inefficient government action.”).
. Id. at 1665 (“The relationship between homeowners, property values, and property taxes at the local level means that requiring local governments to pay just compensation will force them to internalize at least some of the costs of their actions.”).
. Id. (“[F]orcing local governments to pay for land use regulations will, in fact, force them to internalize both the costs and benefits of their actions. The close relationship between property taxes and property values—whether use value or market value—places both the burdens and benefits of government actions on local homeowners.”).
. See Serkin, Big Differences, supra note 15, at 1648 (suggesting that homeowners’ incentive to control local politics is driven by their desire to preserve the value of their property); see also Fischel, supra note 27, at 88–89 (stating that homeowners pay close attention to local policies because they affect the value of their property).
. See id. at 1666 (stating that local governments are more risk averse than larger governments); see also Christopher Serkin, Insuring Takings Claims, 111 Nw. U. L. Rev. 75, 82–88 (2016) [hereinafter Serkin, Insuring Takings Claims] (arguing against the theoretical claim that local governments are risk neutral because they can always raise taxes).
. 304 N.W.2d 455 (Mich. 1981), overruled by Cty. of Wayne v. Hathcock, 684 N.W.2d 765, 787 (Mich. 2004); see William A. Fischel, The Political Economy of Public Use in Poletown: How Federal Grants Encourage Excessive Use of Eminent Domain, 2004 Mich. St. L. Rev. 929, 943 (2005) [hereinafter Fischel, Political Economy].
. 304 N.W.2d at 457. Professor Berger appropriately distinguishes eminent domain from regulatory takings on the grounds that the government actually ends up with title to the real property in eminent domain. Berger, supra note 1, at 12–13. That distinction is irrelevant to observations here about the distorting effects on government incentives based upon the sources of municipal revenue.
. See Distribution of the Burden of New York City’s Property Tax, Furman Ctr. for Real Est. and Urb. Pol’y (2011), https://furmancenter.org/files/publications/ Distribution_of_the_Burden_of_New_York_Citys_Property_Tax_11.pdf [https://perma.cc/K4EZ-PNR2] [hereinafter Distribution of the Burden] (explaining the range of tax payment in New York City); see also Shifting the Burden: Examining the Undertaxation of Some of the Most Valuable Properties in New York City, Furman Ctr. for Real Est. and Urb. Pol’y (July 2013), https://furmancenter.org/files/ FurmanCenter_ShiftingtheBurden.pdf [https://perma.cc/7HY6-25WC] (describing persistent under-taxation of some of New York City’s most valuable property); Ronda Kaysen, Property Tax Envy: Property Taxes, Tax Abatements, Affordable Housing, N.Y. Times: Ask Real Est. (Oct. 25, 2014), https://www.nytimes.com/2014/10/26/realestate/ property-taxes-tax-abatements-affordable-housing.html (comparing apartments and larger, multifamily buildings that are assessed at 45% of their value to single-family homes or properties with no more than three units that are assessed at 6% of their value).
. See Distribution of the Burden, supra note 35, at 8 (“Although the burden of the property tax is spread widely, it is not distributed equally . . . New York City’s property tax system provides for radically different tax treatment of equally valuable properties, depending on the use of the property and the form in which it is owned.”).
. See, e.g., William K. Jaeger, The Effects of Land-Use Regulations on Property Values, 36 Envtl. L. 105, 106, 112–17 (2006) (discussing effects of zoning and land use controls on property values).
. See, e.g., Abraham Bell, Not Just Compensation, 13 J. Contemp. Legal Issues 29, 33 (2003) (describing the moral hazard problem); Blume & Rubinfeld, supra note 2, at 577; Steve Calandrillo, Eminent Domain Economics: Should “Just Compensation” Be Abolished, and Would “Takings Insurance” Work Instead?, 64 Ohio St. L.J. 451, 493–94 (2003) (summarizing the treatment of the incentive issue in economic literature); Eric Kades, Avoiding Takings “Accidents”: A Tort Perspective on Takings Law,” 28 U. Rich. L. Rev. 1235, 1242–43 (1994) (comparing the taking clause and basic casualty insurance); Louis Kaplow, An Economic Analysis of Legal Transitions, 99 Harv. L. Rev. 509, 532 n.61 (1986) (identifying the problem of investment incentives); Thomas Miceli & Kathleen Segerson, Regulatory Takings: When Should Compensation Be Paid?, 23 J. Legal Stud. 749, 756 (1994).
. See Michael A. Heller & James E. Krier, Deterrence and Distribution in the Law of Takings, 112 Harv. L. Rev. 997, 998 (1999) (highlighting Phillips v. Washington Legal Foundation and the Supreme Court’s failure to shed light on the Constitution’s “just compensation” clause).
. For a leading account of moral hazard, see generally Tom Baker, On the Genealogy of Moral Hazard, 75 Tex. L. Rev. 237 (1996), which challenges the traditional ideas of moral hazard defined as “the tendency for insurance against loss to reduce incentives to prevent or minimize the cost of loss.”
. Versions of this example are ubiquitous in the literature. Some examples include: Lawrence Blume, et al., The Taking of Land: When Should Compensation Be Paid?, 99 Q.J. Econ. 71, 73–74 (1984); Blume & Rubinfeld, supra note 2, at 586; David A. Dana, Natural Preservation and the Race to Develop, 143 U. Pa. L. Rev. 655, 678 (1995); Miceli & Segerson, supra note 40, at 756–57.
. Blume & Rubinfeld, supra note 2, at 618 (“Knowing that full insurance is available, private investors will tend to put too much of their physical capital on ‘risky’ land, and too little on land that will be unaffected by the government decision.”).
. Id. (“[In a stylized jurisdiction with ten citizens,] consider a more likely case: the proposed regulation will benefit each of citizens 1–6 (the members of the minimum winning coalition) by $2000, while costing each of citizens 7–10 $4000. Citizens 1–6 will support the regulation because it leaves them better off by $400 ($2000 in direct benefits minus one-tenth of $16,000, or $1600, in compensation taxes), and the inefficient regulation will therefore pass. We should expect most proposed regulations with redistributive effects to resemble this example, for supermajority coalitions will always have the incentive to expel all members in excess of a bare majority.”).
. Cf. Span, supra note 17, at 22–23 (arguing that “[u]nless compensation is made to come out of the personal assets of government officials, it is unclear why having to pay it would act as a deterrent”).
. Id. at 350 (“If government does not fully internalize the costs of takings unless it must spend its revenues to pay compensation, then why should we expect government to fully internalize the benefits of takings when it does not receive them in the form of revenues?”).
. Cf. id. at 350 (“[T]he takings literature mysteriously assumes that government policymakers discount social costs that are not translated into budgetary expenditures, but do not similarly discount the social benefits derived from government programs. . . . One searches the takings literature in vain for some explanation of this puzzling asymmetry.”).
. But see Michael H. Schill, Intergovernmental Takings and Just Compensation: A Question of Federalism, 137 U. Pa. L. Rev. 829, 860 n.116 (1989) (“[S]everal empirical studies indicate that governments overestimate benefits and underestimate costs generated by public programs.”).
. See Abraham Bell & Gideon Parchomovsky, Givings, 111 Yale L.J. 547, 549, 590 (2001) (claiming beneficiaries of givings, or government distribution of property, should compensate the government when deemed “chargeable”); see also Daniel D. Barnhizer, Givings Recapture: Funding Public Acquisition of Private Property Interests on the Coasts, 27 Harv. Envtl. L. Rev. 295, 298 (2003) (proposing that government monetize benefits conferred on coastal property through government action to use as a credit for future acquisition or condemnation of that property).
. Levinson, supra note 2, at 350 n.12 (“Government sometimes does extract payments from the beneficiaries of its programs through special assessments and user fees, but probably not often enough for cash returns to be viewed as a substantial factor motivating government behavior on the benefits side.”); see also Saul Levmore, Just Compensation and Just Politics, 22 Conn. L. Rev. 285, 290–93 (1990) (describing the various tools that governments can use to finance their projects).
. See Vicki Been & Joel C. Beauvais, The Global Fifth Amendment? NAFTA’s Investment Protections and the Misguided Quest for an International “Regulatory Takings” Doctrine, 78 N.Y.U. L. Rev. 30, 97 (2003) (“If, for example, a local government passes a land use restriction that benefits neighboring land by either protecting or increasing its value, the government will capture at least some of that avoided loss (or gain) through property taxes.”); see also Serkin, Big Differences, supra note 15, at 1646 (“Local governments therefore internalize costs and benefits through property values and property taxes.”).
. Id. at 14 (“The cost experienced by the government is, of course, not the same as that experiences by the property owner. But neither is the benefit. Both the cost and the benefit reflect the percentage of value reached by taxation . . . .”).
. See Serkin, Insuring Takings Claims, supra note 30, at 83–84 (illustrating that local municipalities may use either debt or insurance interchangeably as a mechanism to deflect sudden spikes in property tax rates); see also Amy v. Puelz & Robert Puelz, Managerial Use of Debt to Fund Municipal Government Risks, 28 Decision Sci. 745, 746 (1997) (noting that the common municipal approach to insurance consists of bonds for moderate losses and formal insurance for extreme circumstances).
. See generally Christopher Serkin, Public Entrenchment Through Private Law: Binding Local Governments, 78 U. Chi. L. Rev. 879, 906 (2011) [hereinafter Serkin, Public Entrenchment] (explaining that the government may use debt to fund various types of large-scale projects).
. See Berger, supra note 1, at 34, 37 (explaining that the requirement for full compensation prevents government from enacting land use regulations because the benefits will not equalize over time).
. See Serkin, Public Entrenchment, supra note 78, at 961 (“A government’s floating a municipal bond to fund infrastructure development . . . allocates to the future the obligation to pay a fair share of the cost of the infrastructure, given the intertemporal dispersion of benefits.”).
. See Serkin, Insuring Takings Claims, supra note 30, at 83 (“Faced with a significant liability—whether a legal judgment or otherwise—a local government could float a bond to cover the cost.”); see also Puelz & Puelz, supra note 77, at 746 (stating that debt is the municipality’s most commonly used alternative to insurance).
. Serkin, Insuring Takings Claims, supra note 30, at 84 (“[S]tate law often caps the amount of debt that a municipality can incur.”); see also Serkin, Public Entrenchment, supra note 78, at 906 (discussing municipal debt limits).
. In the late nineteenth century, local governments engaged in a kind of spending arms race to try to attract railroads. While this worked in some instances, and created a new economic engine, many municipalities lost out after already investing vast sums of money for economic gains that never appeared. See Serkin, Public Entrenchment, supra note 78, at 906–07.
. See Serkin, Insuring Takings Claims, supra note 30, at 84 (“Just as insurance transforms the risk of spiky losses into more predictable but certain ex ante costs, debt does the same thing ex post.”).
. See id. at 93–100 (describing typical municipal insurance coverage); cf. John Rappaport, How Private Insurers Regulate Public Police, 130 Harv. L. Rev. 1539, 1552–54 (2017) (positing reasons why a government, although “risk neutral,” might choose to purchase municipal liability insurance).
. Id. at 107 (defining inverse condemnation as “an eminent domain proceeding triggered by the property owner instead of the government” that acts as a “mechanism for vindicating a regulatory takings claim in state court”); see id. at 105–07 (discussing inverse condemnation exclusion).
. Id. at 124 (“There is no greater moral hazard problem associated with regulatory takings than with other insurable municipal risks, like police misconduct, employment discrimination, or automobile accidents.”).
. Id. at 110–11; see also Serkin, Big Differences, supra note 15, at 1665 (“Because local governments discount the benefits and put a premium on the costs of their actions, they will tend to underregulate as compared to risk-neutral governments.”).
. See James E. Krier & Stewart E. Sterk, An Empirical Study of Implicit Takings, 58 Wm. & Mary L. Rev. 35, 54–55, 88 (2016) (summarizing findings and concluding “our survey demonstrates that in state court practice, relegation to ad hoc adjudication has marked the death knell for a takings claim”).
. See Serkin, Insuring Takings Claims, supra note 30, at 78 (“Just the litigation costs alone of defending land use regulations can be exorbitant.”); see also Ellen S. Pryor, The Tort Liability Regime and the Duty to Defend, 58 Md. L. Rev. 1, 6 (1999) (suggesting that up to 41% of insurance costs are derived from defending against a claim).
. See generally Serkin, Passive Takings, supra note 15, at 346 (arguing that the government’s failure to regulate to protect property owners from dangers interfering with their reasonable expectations of land may be unconstitutional and could require compensation).
. See id. at 357–58 (citing Frank I. Michelman, The Supreme Court, 1968 Term—Foreword: On Protecting the Poor Through the Fourteenth Amendment, 83 Harv. L. Rev. 7, 9 (1969) (arguing that the government has an affirmative duty to protect the poor from dangers inherent to social inequality)); see also Charles A. Reich, The New Property, 73 Yale L.J. 733, 785 (1964) (asserting that individuals should have a right to certain public benefits such as unemployment and insurance for the elderly).
. See, e.g., Litz v. Md. Dep’t of Env’t, 131 A.3d 923, 931 (Md. 2016) (finding as a matter of Maryland law that an inverse condemnation claim may stand based on government’s failure to act where it had an affirmative duty to act).
. Id. at 363 (“[A]voiding liability is—aside from the baseline—conceptually indistinguishable from receiving payment. Forcing the government to pay for forgone benefits when it fails to act (i.e., passive takings liability) therefore amounts to allowing the government to recover payment for the benefits of its actions.”).