Working paper

Protection Against Government Takings: Compensation for Regulation? (WP 02/18)

Author: Kevin Guerin


The paper responds to recent debate in New Zealand on the power of the government to take private property, directly or through regulatory constraints. This aspect of regulation has received less attention in New Zealand than it warrants. This paper addresses the issue of which protections against takings are appropriate, and the role of compensation as a protective device. A taking can be broadly defined an act by which a government assumes or assigns control over all or part of a property right held by a private party. Government regulation is typically not treated as a taking. In practice, compensation is normally required only for physical takings, such as the acquisition of land, and is not available for takings through regulation, such as restricting the right to use land in a particular way. New Zealand has three options for improved protection against takings: a tighter regime for scrutinising the quality of regulation, more restricted takings powers, and extended compensation provisions. The desirability and practicality of a greater role for compensation requires, however, more detailed consideration. The paper aims to stimulate further debate in this area as an aspect of the wider debate on regulatory quality.


Comments from colleagues have been much appreciated. Particular thanks also to Bryce Wilkinson for his extensive and highly constructive suggestions.


The views expressed in this Working Paper are those of the author(s) and do not necessarily reflect the views of the New Zealand Treasury. The paper is presented not as policy, but with a view to inform and stimulate wider debate.

1  Introduction#

This paper explores compulsory state acquisition of private property rights, primarily by reviewing existing literature. It consciously surveys a range of international economic and legal perspectives, rather than a focus on the New Zealand status quo. It does discuss the current New Zealand approach, but is not intended to represent a comprehensive summation of the New Zealand constitutional position or legal practice. The analysis is intended to advance debate on these concepts in the context of a wider context on improving regulatory quality in New Zealand.

The takings topic has been discussed recently in the review of the Public Works Act, and in Wilkinson (2001) where examples cited of possible New Zealand takings included compulsory acquisition of land, statutory revision of Maori reserve land leases, or even a requirement to provide statistical data for free. This takings aspect of regulation is the focus of this paper because of its topicality, the relative lack of coverage of it in the New Zealand context, and the distinct issues it raises regarding the nature of property rights.

A taking can be seen as simply a subset of the general concept of government regulation, or as one of a range of regulatory instruments available to government to achieve its purpose (albeit one that requires reassigning a property right).[1] Regulation, however, may not necessarily be treated as a taking unless it causes disproportionate private loss or sharply changes property rights expectations. The wider issue of improving regulatory quality overall to improve cost-benefit trade-offs and reduce compliance costs is touched on in section 6, but is not the focus of this paper.

The central question underlying the paper is “what protections against takings are appropriate, and specifically what is the role of compensation?”[2] Addressing this question requires defining a taking. A taking is broadly the act by which government assumes or assigns control over all or part of a property right (or legal right) held by a private party. It can include “appropriating for one's own or another's use or benefit”, or “assuming occupancy of” property.[3] In economic terms, it can be seen as government appropriating private property to achieve a public benefit.

The most longstanding principle underlying takings is “eminent domain”. This is defined as “sovereign control over all property in a state, with the right of expropriation” (Concise Oxford Dictionary, 1995) or the right of a government to appropriate private property for public use, usually with compensation to the owner.[4] Eminent domain is typified by compulsory acquisition of land, or rights to enter land, for public works. There is little if any dispute that when the government deprives someone of physical property, usually land, a taking has occurred.

“Physical” takings can be defined as where government or its agent actually assumes full ownership and/or occupation of a property right or a distinct portion of a property right. “Regulatory”or partial takings occur where government or its agent limits the nature of a property right by means of legislation, regulation, planning processes, permits or other regulatory means. Whether this distinction is meaningful is discussed in Section 4. A common example of a regulatory taking would be a planning requirement that limited the use to which property could be put, or the statute that authorises such requirements. A subset occurs where unrelated conditions are attached to a permit for a specific property use.[5] Whether such requirements are of general application (thereby both spreading costs and benefits) or single out specific property (more likely to impose a disproportionate cost) can also affect perceptions of acceptability and compensability.

Takings operate within a system of property rights supported by a rule of law and a constitutional framework.[6] This dates back as far as the Magna Carta.[7] The manner in which takings take place must be consistent with that system.[8]

Sections 2, 3 and 4 of this paper primarily review existing literature on definition of a taking, justifications for carrying out a taking, when compensation should apply and at what level, and the distinction between physical and regulatory takings. New Zealand’s use of takings is briefly outlined in Section 5. Section 6 discusses non-compensatory means of improving incentives on regulators. Concluding remarks are in Section 7.


  • [1]In this sense, takings should be subject to the same regulatory quality regime as other forms of regulatory activity and considered in the context of the role of institutions and instruments in regulatory design, the focus of the work programme in which this paper falls.
  • [2]Epstein (1985) notes that “all arrangements between the state and individuals are broken down into a network of relationships between different individuals” and from this approach identifies 4 questions for a takings agenda:
  • [3]New Zealand guidelines on legislation refer to alteration of vested rights and ‘the principle that property will not be expropriated without full compensation” (Legislation Advisory Committee, 2001), while Standing Order 382 (2) (b) allows the Regulation Review Select Committee of Parliament to draw Parliament’s special attention to a regulation that “trespasses unduly on personal rights and liberties” (House of Representatives, 1996).
  • [4]“The property of subjects is so far under the eminent control of the State, that the State, or the sovereign who represents it, can use that property, or destroy it, or alienate it, not only in cases of extreme necessity, which sometimes allow individuals the liberty of infringing upon the property of others, but on all occasions, where the public good is concerned, to which the original framers of society intended that private interests should give way. But when that is the case, it is to be observed, the state is bound to repair the loss of individuals at the public expense, in aid of which the sufferers have contributed their due proportion” (Grotius, 1625).
  • [5]This can be inefficient where the value to the public of the condition (obtained at no explicit cost) is less than the cost imposed on the property right owner (Wilkinson 2001).
  • [6]New Zealand law includes, by virtue of the Imperial Laws Application Act 1988, ancient statutes securing liberty and due process including Magna Carta and the Bill of Rights 1688, and the common law of England (including the principles and rules of equity), so far as it was part of the laws of New Zealand immediately before the commencement of the Act. The New Zealand Bill of Rights Act 1990 is also intended to affirm, protect, and promote human rights and fundamental freedoms in New Zealand.
  • [7]The Magna Carta states that “No free man shall be seized or imprisoned, or stripped of his rights or possessions, or outlawed or exiled, or deprived of his standing in any other way, nor will we proceed with force against him, or send others to do so, except by the lawful judgement of his equals or by the law of the land” (Davis, 1989).
  • [8]In the New Zealand context, guidelines note that “this body of rules imposes restraints on individuals and groups within society and regulates the way they exercise various freedoms. But at the same time it both confers and protects important rights, liberties and benefits. As a system it works only if the great majority of society and all major sections within it see it as supporting and protecting their interests.” (Legislation Advisory Committee, 2001).

2  The case for and against takings#

Arguments for and against takings centre on the most basic principles of private rights and public needs. The argument for allowing state takings is that in their absence a significant public benefit would be frustrated. This requires that the activity cannot be progressed without use of that right, it is unavailable without state coercion and there is a net social benefit from the activity.

That unavailability could be a choice by owners of property rights. That choice could be because owners do not want to part with the right or, because they are attempting to capture such a high share of the benefits that the project would not be viable. The latter case represents the “hold out” problem.

Alternatively, the right could be unavailable because the transaction costs of arranging the transaction are so high as to preclude a voluntary exchange – this could be due to a market failure such as imperfect competition or imperfect information (Miceli and Segerson, 1998), or simply due to the number of people involved. The economic concept of a public good is relevant here, as high transaction costs or the free rider problem could limit provision.[9]

A classic example of a taking is where a road must follow a particular route. The state, if negotiations fail, acts to acquire the land by compulsion but compensates the land owner for their loss. Why compensation is paid, and how much is due, will be discussed later in this paper.

The use of eminent domain by private entities to provide public services creates concern, as the powers of the state are being used for the benefit of private parties. Existing utilities, however, have used these powers to establish their networks (at which time they may have been state entities, as they were in New Zealand) and may depend on them for continued access, although possibly as a last resort given statutory requirements for using them and the potential for controversy. There may also be a public interest in continuing such powers to allow existing firms to maintain network stability and so new firms can enter the market (Wilkinson, 2001); ie, for competition reasons. Roads and public utilities also carry a degree of public benefit regardless of who is operating them.

Public interest can also be a flexible concept with the power of eminent domain used for commercial development that is argued to be in the public interest. This is a common issue in the US:[10] for example, in two prominent cases “New York City also recently announced its plan to condemn a block in the Times Square area for a new building for the New York Times[11]“ and a similar plan has been announced for the New York Stock Exchange.[12]

This wider interpretation of the public interest raises obvious concerns, but these can be considered within the same overall framework as what are clearly public works; ie, has a taking occurred and is compensation required? A public interest restriction on takings of land, if practical, would resolve this issue in that specific context, but whether government powers in general should be used for private interests is a wider issue not within the scope of this paper.[13]

A pure approach could suggest that eminent domain should only be applied to hold out situations where it is a necessity, there is public access to the resulting activity and the resulting surplus is shared proportionally between the parties to a forced exchange. On this basis, the essentiality of an activity and the inability to achieve a voluntary outcome might represent a form of public benefit test that could be practically implemented (Wilkinson, 2001).

The existence of a takings power assumes that there are circumstances where the public benefit of over-riding property rights exceeds the public and private benefits of protecting those rights. Giving the state, or its agents, any ability to carry out takings still, however, carries risks. Arguments for limiting the ability of governments to carry out takings usually focus on two main aspects, constitutionality (common law, natural rights and liberty, democracy) and efficiency (Wilkinson, 2001 and Epstein, 1999).

The primary means of avoiding abuse of the takings power are to limit its scope, constrain the processes through which it can be exercised, and to require payment of compensation. The first two points are addressed later. The latter is the subject of the next section.


  • [9]A pure “public good” is a good that is non-rival in consumption (so your consumption does not affect another person’s ability to consume it) and is non-excludable (so you can’t stop another person consuming it); eg, national defence. These characteristics can create a “free rider” problem where potential consumers have an incentive to hide their true willingness to pay since they will still be able to consume the good. If this effect is strong, public goods will be under-provided. When a free rider problem is addressed through coercion, such as a taking, it can create a “forced rider” problem where some consumers bear a cost of provision that exceeds the benefits they receive.
  • [10]US federal and state constitutional provisions usually refer to “public use”. This has been extended to slum redevelopment, and increasingly for general economic development purposes including factories, stores, or parking lots where the public purpose may be additional tax revenue . In some US cases, “courts held that the condemnation orders were serving an improper -- usually private – purpose” such as “a retail redevelopment” (Keeney, 2002).
  • [11] and .
  • [12]“New York City Economic Development, with aid from its state sister agency, has decided to condemn one of lower Manhattan’s earliest skyscrapers and tear down the historic structure to make way for the exchange’s new home, a 900-foot-tall steel-and-glass tower” .
  • [13]The specific issue of compulsory acquisition of land is governed in a New Zealand context by the Public Works Act 1981. This Act is currently under review (Land Information New Zealand (2000b and 2001).

3  Compensation#

It is impossible to address the concept of a taking without being drawn into debates (with deep constitutional implications) over when it is desirable (efficient) to pay compensation, at what level it should be set and who should be liable to pay it. This section discusses these questions, through a review of literature, and without distinguishing between physical and regulatory takings. The nature and implications of such a distinction are considered in the next section.

3.1  Why compensate?#

Compensation for takings is to a large degree about incentives. A prime element of this is the problem of “fiscal illusion,”[14] where the true cost of the proposal is hidden. The Italian scholar Amilcare Puviani in 1903 in The Theory of Fiscal Illusion (Buchanan, 1987) addressed fiscal illusion through the question “If the ruling group desires to minimize taxpayer resistance for any given level of revenues collected, how will it set out to organize the fiscal system?”[15] Takings, which carry no explicit fiscal cost or a reduced cost, can become the preferred means of achieving a policy objective, through reducing the apparent cost, even if the total social cost actually exceeds the total social benefit.

A second argument relates to incentives for private parties to encourage use of the takings mechanism rather than engaging in a voluntary transaction at their own cost. Issues around the advantages of making beneficiaries pay compensation and of requiring proportional sharing of the benefits of a taking are discussed in later sections. Finally, there is the longer-term incentives issue, dealing with implications of takings over time for social acceptance of the legitimacy of state action, and for the value of privately held property rights.

There has been an extensive debate in economics literature about the implications for efficiency and national welfare of paying or not paying compensation for takings. The following discussion and Table 1 attempt to summarise the key points of that debate.

Compensation has historically been argued to be due when “a given person has been required to give up property rights beyond his just share of the cost of government” (Stoebuck, 1972). This principle is not usually applied to taxation, possibly because the burden is widely spread and general tax decisions made by the legislature have greater social acceptance than bureaucratic or executive action to take specific property rights from specific individuals. The latter may also offer more scope for abuse or victimisation.

It has been argued that compensation for the use of “eminent domain” and “no taxation without consent” are fundamental rules limiting the power of government and crucial to the consent of the governed (Wilkinson, 2001). This is a constitutional debate.[16] More prosaically, compensation mimics the consideration paid in a voluntary exchange and willingness to pay it may show that the public benefit exceeds the cost. Payment of compensation can also separate the social benefits of the taking from the income distribution effects, reducing any unfair redistributive effects (Blume and Rubinfeld, 1984).

Arguments in specific cases for not compensating include (Wilkinson, 2001) where the costs of identifying and assessing rights to compensation exceeds the benefits of paying it, the benefits affect the same people as the costs, and the taking is restricting the use of market power to the risk adjusted return on capital. It has also been argued, from an economic rather than constitutional perspective, that full compensation creates moral hazard. Investors seek to raise the cost of compensation to discourage a project, knowing that at worst they will be reimbursed, while landowners ignore the social costs created by their inefficient investment if a taking occurs.

One response to this is that under-compensating creates an incentive for takings rather than voluntary exchange (due to fiscal illusion). Another is that not all the cost of an investment (Wilkinson, 2001) or all the values owners attach to their land (Miceli and Segerson, 1998) will necessarily be reflected in market value. The latter problem is not easy to address in practical terms as owners have an incentive to over-state non-market values.[17] It can also be argued that a lack of compensation encourages over-investment if it is perceived that the government is less likely to take valuable land. There is evidence that governments over-compensate for or avoid taking high-value land, and under-compensate for and therefore accumulate excessively, low-value land (Munch, 1976).

Compensation can also be seen as a form of public insurance against a taking (Miceli and Segerson, 1998) particularly if the level of compensation is uncertain. This insurance can be argued to be offset by any tendency of risk averse owners to accept lower offers to avoid litigation costs and by fiscal illusion by governments. Limited counters to these problems include fiscal constraints by government and a possible rule, which would impose all legal costs on the government if it lost (Eposto, 1996).

It has been argued that compensation is not required where the benefits affect the same people as the costs (Wilkinson, 2001). This was covered by Justice Holmes in the concept of “average reciprocity of advantage” where “the share of each party in the benefit …. Is sufficient compensation for the correlative burden (Brauneis, 1996). Regulation that singles out specific property owners without compensation would not meet this equal protection test – but that would not necessarily mean it was inefficient.[18]

The above arguments are primarily subjective and open to differences in interpretation. A number of attempts have been made to design explicit rules to avoid or mitigate the incentive problems associated with compensation. Since these rules are variations on a theme of achieving an efficient outcome, but the variations reveal interesting subtleties about the impact of different objectives and of governmental behaviour, they are summarised together in Table 1 rather than discussed at length in the text.

In terms of the central question underlying this paper “what protections against takings are appropriate, and specifically what is the role of compensation”, the rules in Table 1 focus sharply on the efficiency of compensation in terms of maximising national welfare.

Considering efficiency requires determining what efficiency standard to apply. The literature has argued (Fischel and Shapiro, 1988) for a variety of standards. Pareto-efficiency would require that no one is left worse off and society is better off. Michelman’s standard is for society to be better off but compensation is paid only if it is less expensive than demoralisation costs. The Kaldor-Hicks criterion requires that projects should proceed if society is better off but compensation does not then have to take place

Whichever efficiency standard is chosen, the application of most of the rules cited above requires assessing the social costs and benefits of the taking, and of the effects of not compensating. These assessments carry significant information burdens, while the latter also involves judgements regarding social perceptions of what constitutes a property right and expectations of how rights will change over time.

- Table 1 Compensation rules
Rule Rule Summary
Utilitarian approach – pay compensation if cheaper than the effects of not compensating Test is fairness.  Linked to Hume’s theory of property, with private property emerging from rules needed to allow association of individuals.  Arbitrary encroachment on private property has serious disvalue.  Bentham’s theory of social utility suggests protection of extant rules on property from capricious redistribution is crucial to gains in productivity.  Reject taking unless efficiency gains (E) outweigh either settlement (S) or demoralization (D) costs:[19]  E > min (D, S).  This does not require Pareto-efficiency.  Undertaking inefficient projects, however, if E > min (D, S), would raise demoralization costs and compel compensation.  (Michelman, 1967). 
Compensate when government acting as an enterprise Compensation when government is acting in an enterprise capacity, but not when it is acting in an arbitral capacity  (Sax, 1964).
No compensation for insurable risks Compensation as insurance against adverse effects of government regulation.  Private insurance limited[20] as is government insurance.[21]  Compensation after the fact avoids these problems, but should only be paid when benefits of doing so exceed the costs.  Focus on risk aversion (inverse to wealth?) and where loss is substantial diminution of value.  Compensate for large losses to those unable to insure against those losses  (Blume and Rubinfeld, 1984).
Lump-sum compensation With lump sum compensation, if projects are appropriately evaluated, investors are risk-neutral and compensation is independent of current land-use, then private investment decisions are socially efficient.  Zero compensation is efficient as a special case.  With full compensation, investment will exceed socially optimal level, particularly if over investment discourages takings.  If fiscal illusion is present, some compensation may be necessary to avoid excessive takings (Blume, Rubinfeld and Shapiro, 1984).
Compensation depends on applicable constitutional model Private insurance is a limited option and does not address demoralisation costs.  Pay compensation or not based on expected government behaviour.  Zero compensation if no reasonable compensation rule deters takings (compensation will then create moral hazard for investors) or if the government is Pigouvian, acting solely on benefit-cost calculus (so no constraint is necessary).  Under majoritarian government, optimal compensation will be more than zero, to discourage excessive takings, but less than full, to minimise moral hazard (Fischel and Shapiro 1988, 1989).
Efficiency For singling out, ex ante efficiency, so landowners are compensated for acting efficiently.  For general regulation, ex post efficiency, so compensate if the regulation is inefficient (Miceli and Segerson, 1994). Full compensation at efficient level of investment as, if landowners consider the value of the land is likely to affect takings decisions, under-compensation will encourage over investment and over-compensation will encourage under investment (Miceli and Segerson, 1998).
Society’s gains If the social benefit is known, pay the owner the social value of the taking for the property, or charge the owner that amount.  If social benefit not known, apply “two-part tariff”.  The owner announces private benefit and a monetary transfer occurs regardless of whether taking proceeds.  The transfer is the social gain, if the taking occurs at a price equal to the private benefit, and may be negative.  The private benefit is paid if the taking occurs.  (Hermalin, 1995). If social benefit is unlikely to be known, this defaults to two-part tariff which would penalise the owner if a taking occurred with a social value less than the private benefit, as can occur when fiscal illusion applies.

The rules in Table 1 therefore generally appear to assume a level of information that in practice is unlikely to exist, making the feasibility of applying this economic approach in real life questionable. In the absence of accurate assessments of social benefit and cost, and of government decisions based solely on such assessments, the balance of judgement from these rules (as distinct from a constitutional context around constraining government) is unavoidably subjective. This applies to both whether a taking should proceed, and whether and how much compensation should be paid. Economics does not give us a hard and fast operational framework for takings compensation.


  • [14]Similar approaches can of course be taken in the design of regulation, and regulation itself can serve to hide costs by transferring them to third parties and rendering them implicit rather than explicit. It is also interesting to compare the hypothesised political goal of minimising taxpayer resistance with the economic goal of minimising the deadweight costs of taxation, or regulation. Where the two goals are aligned, so perceived costs equal real costs, presumably the outcome will be the socially optimal one, while otherwise it will be suboptimal.
  • [15]Mechanisms suggested included distorting the link between the total cost of a programme and the individual share, charging taxes at favourable times such as when an inheritance has just been received, charging fees for nominal services such as licences associated with positive events such as marriage, linking taxes to public opinion such as surcharges on business profits, using scare tactics, or introducing taxes where the ultimate incidence is unclear (Buchanan, 1987).
  • [16]Highlighted in the New Zealand context, by references to vested rights and compensation in the Guidelines (Legislation Advisory Committee, 2001) cited in the Cabinet Manual (Cabinet Office).
  • [17]Michelman notes that courts usually award either nil or just compensation so that legislated settlement rules may be appropriate, and cites the example of relocation payments for households displaced by federally financed redevelopment activities (Michelman, 1967). The New Zealand parallel is the solatium payment, separate from the calculation of market value, under the Public Works Act where residential property is taken. The New Zealand legislation is otherwise based on market value, an administratively simple approach, and on making those affected no better or worse off than under a voluntary exchange.
  • [18]In Michelman’s terms the risks of abandoning efficient projects due to stringent compensation, or unduly concentrating losses due to les stringent compensation are minimised by insisting on compensation “when settlement costs are low, when efficiency gains are dubious and when the harm concentrated on one individual is unusually great”, and relaxing that insistence “when there are visible reciprocities of burden and benefit, or when burdens similar to that for which compensation is denied are concomitantly imposed on many other people” (Michelman, 1967).
  • [19]Demoralisation costs are losses from realising there will be no compensation, and future lost production caused by fear of similar treatment. These costs are distinct from random risks such as natural disasters, and arise from cases where low settlement costs would have made compensation easy, burdens are seen as disproportionate, inefficient projects are seen as vehicles for redistribution, there are no reciprocal benefits, or there is little confidence of benefit from future projects (Michelman, 1967). A Pigouvian government acting on benefit-cost calculus would generate more random risks and smaller demoralisation costs (Fischel and Shapiro,1988).
  • [20]The main problems are measurement difficulties, the high risk of moral hazard and adverse selection (landowners may be able to affect the taking decision and those at high risk of a taking are more likely to insure), inability to secure insurance company assets against taking, and the fact that demoralisation costs occur when the effect of the taking is realised by land owners, not at the time of the legal taking itself (Fischel and Shapiro, 1989).
  • [21]Government would have incentives to take uninsured land, distorting decisions on whether to insure, and setting premiums for low-probability events is difficult (Blume and Rubinfeld, 1984).

3.2  Who should pay compensation#

The question of whether compensation should be paid cannot easily be separated from the question of who should pay it. A crucial factor here is the incentives that are created by a liability to pay compensation.

Incentives on government, taxpayers, lobbyists and investors are all crucial to the takings issue. When compensation is funded by taxpayers, there is likely to be less effective scrutiny of cost and a greater incentive to achieve the goal by regulatory means rather than an explicit taking, although on the plus side there is a chance that greater transparency will generate an improved outcome.

Requiring beneficiaries to pay the costs of a taking can in theory create better incentives even when compensation is not due (Wilkinson, 2001). When compensation is due, and the beneficiaries of a taking are clearly identifiable, there is a strong argument in both efficiency and equity terms for the compensation to be paid directly by the beneficiaries.

This argument can be extended. As noted above, principles such as compensation for eminent domain can be crucial to maintaining the consent of citizens that underlies effective government. Consent itself, of course, can be debateable when it relates to redistributive actions of government in a majoritarian system.

This has led to suggestions such as supra-majorities for non-proportional taxes, funding redistributions from within those who voted for them or applying public interest tests to the purposes of taxation (Wilkinson, 2001). Another example is Epstein’s (1985) approach of proportional distribution of the benefits of a taking among the parties to the forced exchange. Where there was a true public benefit and therefore a net surplus, this would result in compensation above market value (Wilkinson, 2001), which could be particularly appealing (in terms of public acceptance of the legitimacy of takings) where private parties were exercising eminent domain.

As with the efficiency of compensation debate reviewed above, issues of practicality would severely constrain such solutions. This does not, however, preclude it being of value in particular cases.

4  The boundaries of takings#

This section reviews issues around when a government action becomes a taking. This is particularly difficult for regulatory takings where the nature of the property right can be crucial to the judgement.

4.1  When is a regulation not a taking?#

Some cases of government appropriation of private property are ordinarily not treated as requiring compensation. The single strongest example is general taxation. This is in theory a taking, as private property is compulsorily acquired. The empowerment of parliamentary representatives to levy such impositions for public benefit, and the fact that both the burden and the benefits are spread across all members of the public effectively providing in-kind compensation, mean however that it is generally accepted. In addition, to compensate other than in kind for losses which arose through taxation would create an absurd circular effect and render government impossible.[22]

Deregulation is not normally considered a taking, although regulation can be. The government has not, by passing a law, created an enforceable contract with the beneficiaries of that law (Wilkinson, 2001). There are nevertheless situations where compensation or transitional arrangements are used to reduce opposition to deregulatory moves. Action by private individuals that creates competitive harm is also not a common law taking. No force has been used to deter customers; rather the competitor has simply “expanded the set of legitimate alternatives open to them” (Epstein, 1985, emphasis in original).

There is, however, a wide range of government actions where a boundary line is much harder to draw (whether in theory here, or in legal terms by the courts). These focus around the nature of private property rights, particularly where the state is either mediating between conflicting existing property rights or acting to eliminate a nuisance – the “police power” (see the next section). On the question of mediation, it can be argued that where it is a question of accommodating conflicting claims on resources, the primary focus should be on cost allocation between the holders of those claims with no necessity for any taxpayer contribution (Sax, 1971). In such cases, the government’s role could be to achieve an outcome that was “fair” between the parties or “in the public interest”, which could differ.

However, this depends on whether the conflict arises from existing rights as previously understood, a reinterpretation of such rights, or the extension of existing rights. One example is where the government is protecting the rights of individuals (eg, disabled people, Miceli and Segerson, 1998) by imposing costs on property owners. Is this an existing right or a new right? What rights are vested and in whom? This may not be universally seen as a situation of mediation.

Epstein (1985) sets as a guiding principle “would the government action be treated as a taking of private property if it had been performed by some private party?” Another approach is to limit property in a takings sense to “every species of interest in land and things of a kind that an owner might transfer to another private person” (Stoebuck, 1972). Neither of these approaches appear to reflect the impermanent nature of property rights; ie, the logical ultimate outcome of a “vested rights” concept is to block any legal change without compensation (discussed below) nor is it clear how rights become vested.[23]

It is, however, a powerful argument that unless the concept of takings is applied to regulation, it is ultimately of no effect, as the government would then choose regulatory takings in preference to physical takings or taxation. The following discussions will attempt to throw some light on this perspective.

4.2  Limits on private property rights#

The above discussion illustrates two key questions about the fundamental nature of property rights;[24] ie, (1) are these positive rights (the right to take an action, eg, to use land) or negative rights (the right to prevent an action being taken, eg, to prevent someone else using land); and (2) what is the scope and permanence of the property right – to what degree can the terms of the right be altered without the consent of the right holder (courts can address disputes within existing law, but that leaves the issue of changes to the law – court interpretations can also change over time)?

Property rights are not absolute and unchanging (Bromley, 1993). Any property right can be seen as held subject to a general understanding of the constraints imposed by the community (expressed through judicial interpretation, statutory definition or direct community/peer pressure) with the knowledge that those constraints evolve over time, but that the right will not be unduly altered without consent or compensation.[25] This definition of course leaves much open to interpretation.[26]

The evolution of property rights itself can be driven by efficiency pressures or by interest groups seeking to modify the regime to their own advantage. This paper does not attempt to determine which effect dominates. Any particular change also may or may not have net benefits, nationally or to specific rights holders.

Alternatively, we can “describe property as the value … left after the inconsistencies between … competing owners” of interrelating property rights are resolved, while recognising that “new conflicts are always arising as a result of a change in … technology, or in public values” (Sax, 1964). These changes can include altered perceptions of what constitutes an externality or spillover effect.


  • [22]Compensation for physical takings, however, involves using general taxes to fund payments to a small group – a much simpler and less circular process, where it could also be argued that the benefits of the public works offset the cost of the taxes for the general public, in the same way as the compensation offsets the negative effect of takings for the small group. This is also an example of where singling out individuals or a small group for regulation makes the action more likely to be seen as a taking.
  • [23]A utilitarian approach to a practical judicial rule could be “that compensation is due only when there has been either (a) a physical occupation or (b) a nearly total destruction of some previously crystallized value which did not originate under clearly speculative or hazardous conditions (Michelman, 1967).
  • [24]Generically, property rights include the right to exclusive possession, the right to use and the right to dispose of the property – possession, use, disposition (Epstein, 1999).
  • [25]A subset of the general problem is when rules change after an investment has been committed based on an existing property right but before it is complete. These transition situations raise the same issues in general as for completed investments.
  • [26]See note 2. Also, in the New Zealand context “the balances in society are constantly changing and the legal rules, therefore, are in need of constant review and adjustment ….. the Government of the day must assume responsibility for assessing changes ….. they will be unlikely to gain broad acceptance until they have been developed through an adequate process, including appropriate consultation. There are also important legal principles relating to fairness and the preservation of individual liberty that need to be complied with if the legislation is to prove acceptable.” (Legislation Advisory Committee, 2001).

4.3  US jurisprudence#

Initial US jurisprudence on takings,[27] focused on whether government action affected the existing contract between citizens and the government under the constitutional provisions prohibiting “ex post facto” laws or laws interfering with existing contracts, both seen as impairing vested rights (see Table 2). This jurisprudence began with the concepts that rights that had vested in individuals under pre-existing positive law could not be removed.[28] This, however, was a juridical dead-end as it ultimately prohibited any new government action (Brauneis, 1996).

The alternative approach adopted was to appeal to the Due Process clause (see Table 2), with a focus on determining solely whether there was a need to prevent injury to the community, in which case action could be justified in terms of the “police power. This approach concluded that laws that aimed at protecting an ideal boundary between owner and community rights were acceptable (Brauneis, 1996).

In this substantive rights or police power tradition, there is no need for compensation when the ideal boundary is being protected as no taking occurs by definition, while if it is not, so that the property use being stopped does not injure the community, then the law is ultra vires and compensation is irrelevant (Brauneis, 1996).

Table 2 - US constitutional provisions relevant to takings
Ex Post Facto Clause and Contracts Clause Section 10.  No state shall enter into any treaty, alliance, or confederation; grant letters of marque and reprisal; coin money; emit bills of credit; make anything but gold and silver coin a tender in payment of debts; pass any bill of attainder, ex post facto law, or law impairing the obligation of contracts, or grant any title of nobility.
The Takings Clause Amendment V No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a grand jury, except in cases arising in the land or naval forces, or in the militia, when in actual service in time of war or public danger; nor shall any person be subject for the same offense to be twice put in jeopardy of life or limb; nor shall be compelled in any criminal case to be a witness against himself, nor be deprived of life, liberty, or property, without due process of law; nor shall private property be taken for public use, without just compensation
The Due Process Clause Amendment XIV (1868) Section 1.  All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the state wherein they reside.  No state shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any state deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws. 

Note: Emphasis not in original.

However, the underlying concept here; ie, “sic utere tuo ut alienum non laedas – use your own property in such a manner as not to injure that of another” was itself deemed to be hollow by Justice Holmes as it was impossible to deduce the structure of existing property rights from the general principle (Brauneis, 1996). The meaning of injury evolves over time and is often case specific.

Justice Holmes then stated in 1922 in Pennsylvania Coal Co v. Mahon[29] that “The general rule at least is, that while property may be regulated to a certain extent, if regulation goes too far it will be recognized as a taking”. This rule, also known as the “diminution of value” test is not, however, very helpful as a general principle other than to suggest that drastic or discontinuous changes in property rights should be avoided. After all, in the same judgement, Justice Holmes also noted that “Government hardly could go on if to some extent values incident to property could not be diminished without paying for every such change in the general law”.

There is an issue here around the treatment of public perceptions in the takings debate, particularly on such basic issues as whether “it is at least as important that they believe their “personal property” is protected from confiscation as that it is in fact protected” (Fisher, 1988). Radical changes in perceived rights may alter the general acceptance of takings law, and thereby have an impact on outcomes in practice, notwithstanding constitutional principles.

A later US Supreme Court case concluded by five votes to four that compensation was due for prohibition of development of two beachfront lots, stating that “…when the owner of real property has been called upon to sacrifice all economically beneficial uses in the name of the common good, that is, to leave his property economically idle, he has suffered a taking”. The Court concluded that title could not be considered to be held “subject to the State’s subsequent decision to eliminate all economically beneficial use” (Lucas v. South Carolina Coastal Council, 1992).

This is similar to Holmes’s predisposition against drastic or discontinuous changes in property rights, and along with permanent physical invasion of property, represents the only types of regulatory action categorically considered a compensable taking by the Supreme Court (unless the regulation prohibits uses “not previously permissible under relevant property and nuisance principles”).[30] In the Lucas case, the Court also noted that the harmful use standard for the police power had evolved to a standard of a policy that provided a widespread public benefit and was generally applied, or substantially advanced legitimate state interests, but that the categorical rule cited above still applied to such takings.

The Court has considered that takings can be compensable when using the police power to stop a public nuisance (depending on the merits of the case) but not if using eminent domain to confer a benefit. It has been suggested as a consequence that if a public nuisance was simply defined as the impedance of a public benefit, no compensation would ever be payable (Wilkinson, 2001).

The above discussion should not be taken to suggest that regulation for the purpose of preventing a nuisance or otherwise protecting public health and safety would always represent a taking. Pennsylvania Coal Co v. Mahon is often cited as an example of a taking. In Plymouth Coal Co. v Pennsylvania (Brauneis, 1996), however, Justice Holmes accepted that requiring coal to be left in place at the edges of a property was directly linked to protecting safety and also provided mine owners with reciprocal advantages offsetting their costs. The action therefore either was not a taking in the first place, or was a taking but contained in-kind compensation so requiring no further redress.

Another recent US case refers an interesting extension of eminent domain, through the instrument of compulsory licensing, suggested in this particular case for electronic publication of newspaper articles to avoid “holes in history” (New York Times v. Tasini, 2001). Compulsory licensing has also been applied in other intellectual property situations; eg, for pharmaceuticals. This specific example of a taking is not examined further here due to limitations of space.

4.4  So when does regulation become a taking?#

In the long run, as property rights evolve (by statute, judicial interpretation, or peer pressure), all the above approaches may resolve to the same argument. This is that what constitutes a nuisance or an unacceptable externality created by use of a property right changes over time. These changes can render what was a legal use (for example, a pig farm now surrounded by primarily residential properties) no longer legal.[31]

Who then in effect has stopped that use, and who is liable for the costs of doing so – the affected neighbours or other beneficiaries of the abatement, the courts or the government?

US jurisprudence is of limited use in providing consistent guidance to addressing the New Zealand situation, even setting aside constitutional differences. Is any regulation of property rights acceptable without compensation (in practice if not constitutionally) as long as physical rights of occupation (where applicable) and some economic use remain to the owner? There does not seem to be any sound principle behind such an approach. A recent US Supreme Court ruling has further clouded the boundaries around planning requirements by allowing government agencies to impose overlapping temporary moratoria on land development without paying compensation to landowners.[32] The picture is not clear.

There can be a conflict between the need for certainty when undertaking an investment and the need for a legal system to accommodate change (this is not of course impossible – a system that values the rule of law does not necessarily prohibit change in the law). Regulatory takings “at its worst … requires an arbitrary freezing of social relations at a particular moment in the constant evolution of institutions” (Bromley, 1993).

This is, of course, an extreme view. Realistically, such evolution must occur, but the manner in which it does so is crucial to the ongoing legitimacy of the resulting pattern of property rights at any particular point of time. Changes that are too sudden, too large or too unfair (eg, unnecessary or not compensated) can affect that legitimacy.

In terms of the central question underlying this paper “what protections against takings are appropriate, and specifically what is the role of compensation?” this discussion highlights the inevitable uncertainty around what is a property right, how it becomes vested in an individual or community, and what constitutes taking such a right.

A practical answer must involve neither paralysis of change nor free rein for regulators operating under fiscal illusion. Rather it would need to provide a compromise between flexibility and stability that provides sufficient certainty in property rights to allow investment, and sufficient ability to adjust for changes in society, technology and the environment (Bromley, 1993). Michelman’s approach of compensating when demoralization costs of a taking exceed the settlement costs may fall into this category, if practical or constitutional (Michelman, 1967). This sets aside the issue of who should pay for any compensation, discussed earlier.

In the interim, a rough and ready approach seems to have evolved in many jurisdictions of (1) accepting that appropriation of physical property must be compensated, unless it occurs as part of generic taxation or overarching changes to property rights structures, (2) defining certain other property rights as similar enough to physical property or as sufficiently tied to a specific individual’s ownership, or as so essential for incentive purposes, that equivalent rules should apply, and (3) leaving all other regulation of property rights include land-use rules to go through generic non-compensated regulatory processes generally intended to avoid arbitrary or discriminatory outcomes.

Does this provide adequate protection against takings? Is the role it provides for compensation appropriate? The latter question is addressed below, and is also relevant to the discussion in Section 3.

4.5  Compensation for regulatory takings?#

Governments usually compensate for taking part of a piece of property but not for reducing the use to which that property can be put. For example if a strip of land were taken to provide a safety strip beside a road it would generally be expected that market value compensation would be paid for that land, but if a regulation were made to ban any new construction on land within a certain distance of a roadway no compensation would normally be paid.

This problem is directly linked to several issues discussed above, particularly the incentives problem when there is a choice between two means of achieving an objective but only one carries an explicit cost (ie, fiscal illusion applies), and the question of how absolute a property right is vested in the current owner. The latter issue arises most frequently in the application of planning controls on land.

The Lucas decision clearly assumes that the diminution of value threshold had been passed but does not completely resolve the issue of why depriving an owner of 99% of economic use should be treated differently (Epstein, 2000).[33] Partial land use restrictions, which usually occur by regulatory means, are never compensated, regardless of the constitutional or economic merits of doing so (Epstein, 2000). The diminution of value threshold also focuses solely on the costs of a regulatory action, ignoring the benefits, and under fiscal illusion could encourage regulations that impose small private costs and therefore fall under the threshold (Miceli and Segerson, 1998).

Discussions of compensation for regulatory takings usually run very quickly into the question of affordability, given the ubiquity of such activities. Although one could presumably argue that that is primarily an issue of fiscal illusion (Wilkinson, 2001) and that ubiquity would be reduced if compensation was due, the pervasiveness of overlapping property rights in modern society and the difficulties of offsetting costs and benefits are very real issues.

It seems logical that if two forms of taking, physical and regulatory, are economically equivalent then the availability of compensation should also be the same. An alternative argument, however is that this only follows logically if there is a single economically efficient compensation rule, but if there is more than one such rule then it is appropriate to choose the best rule for the situation (Hermalin, 1995). It is unlikely, however, that the current distinction between compensation for physical and regulatory takings is based on such an evaluation.

Judicial review of compensation in eminent domain cases can be seen as an example of a governance mechanism operating to restrain governmental action, both in terms of cost on its own and uncertainty regarding both cost and feasibility of a taking (Eposto, 1996). At an extreme, a requirement to pay compensation for all takings (or none) could therefore be seen as a means to constrain (or not) the size of government overall.

Wilkinson (2001) argues strongly in the New Zealand context, on efficiency and constitutional grounds, that “establishing the principle of compensation for regulatory takings would have a potentially material and beneficial affect (sic) on regulatory behaviour”, imposing greater discipline to design regimes properly up front, and reduce the demand for regulation by those otherwise able to avoid the associated costs.

The vagueness and evolving nature of property rights, however, seems to still be a major conceptual barrier to wholesale adoption of compensation for regulatory takings, even without considering issues such as fiscal or transaction costs.[34]

4.6  The international dimension#

An example not discussed above of how property rights can be reduced without any taking being explicitly acknowledged or made known is where a government enters into international agreements. These can become binding on individuals with little or no public or parliamentary consideration (unless ratification is legally required which can vary by country and by type of agreement). A variant of this is where trans-national regulatory bodies are created which may be less subject to domestic constraints on regulatory activities (Guerin 2001 and 2002).

Conversely, cross-border investment may be treated better with regard to takings than purely domestic investment in some circumstances.[35] This is because bilateral and multilateral agreements can deal with circumstances in which the value of an investment is reduced by regulatory action of the host government. Provisions can include defining the type of action involved, setting compensation amounts and providing for third-party arbitration. This could result in some differences in the treatment of a particular taking with regard to the different private parties affected. This would not be the case where an agreement simply specified a national treatment or non-discrimination rule.


  • [27]The discussion below draws heavily on United States jurisprudence and literature which is extensive on this subject. It is, however, constrained by a focus on the implications of the wording used in the US constitution, rather than wider economic or social analysis. This needs to be taken into account when drawing wider implications from US outcomes.
  • [28]This jurisprudence was not based on the Takings Clause because it was considered to apply to the federal government only. This was under the view that the Fourteenth Amendment, which applied to the state governments, did not incorporate the first eight amendments. This view was later reversed (Brauneis, 1996).
  • [29]This case involved a law which prohibited underground mining that caused subsidence. The mining company had sold surface rights while retaining mining rights and obtaining a waiver of rights to compensation for subsidence It claimed that the Act deprived it of its property (the waiver) without compensation. The Supreme Court agreed (Pennsylvania Coal Co v. Mahon, 1922).
  • [30]This is very similar to the statement by Holmes that “To make it commercially impracticable to mine certain coal has very nearly the same effect for constitutional purposes as appropriating or destroying it”. That judgement did not preclude regulation of mining, however, as it noted that an existing law which required “a pillar of coal to be left along the line of adjoining property secured an average reciprocity of advantage that has been recognized as a justification of various laws” (Pennsylvania Coal Co v. Mahon, 1922).
  • [31]The US Supreme Court in Village of Euclid v Ambler Realty Co in 1962 upheld a zoning change which banned a factory and caused a 75 percent drop in value, basing the decision on suppression of a nuisance (Epstein, 1999).
  • [32]Property owners argued a sequence of development bans lasting 32 months effectively created a government "taking" of the economic benefit they had expected to derive from the land. The majority noted land use regulations "are ubiquitous and most of them impact property values in some tangential way--often in completed unanticipated ways. Treating them all as . . . takings would transform government regulation into a luxury few governments could afford." The minority wrote "as is the case with most governmental action that furthers the public interest, the Constitution requires that the costs and burdens be borne by the public at large, not by a few targeted citizens." (Tahoe-Sierra Preservation v. Tahoe Regional Planning Agency, 2002). Washington Post, 23 April 2002.
  • [33]Justice Scalia noted in Lucas v. South Carolina Coastal Council that this “is no more strange than the gross disparity between the landowner whose premises are taken for a highway (who recovers in full) and the landowner whose property is reduced to 5% of its former value by the highway (who recovers nothing). Takings law is full of these “all-or-nothing” situations”.
  • [34]There may, however, be more scope to undertake procedural changes designed to limit the potential for undesirable takings to occur in the first place; eg by limiting the ability of regulators to attach discretionary conditions to permits (Wilkinson, 2001).
  • [35]An example is NAFTA (1994 North American Free Trade Agreement) where arbitration panels were established under an anti-expropriation clause, which in practice “could end up giving them protections against regulations far beyond those domestic companies enjoy in their own courts” (Magnusson, 2002).

5  New Zealand application of the takings concept#

The differences in constitutional structure from the US, where much of the takings literature has developed, will complicate application of its lessons to the New Zealand situation. These differences include the lack of a written constitution in New Zealand and the limited judicial constraints on legislative action in New Zealand compared to the US.

The same issues of conflicting property rights and the balance between public and private interest arise in New Zealand, however, as elsewhere, as does a focus on physical occupation of land as the primary arena for the takings debate.[36]

New Zealand’s legal framework is based on a common law heritage. For example, fundamental common law principles require that property will not be expropriated without full compensation (Legislation Advisory Committee, 2001). The checklist in the Legislation Advisory Committee Guidelines, under the heading of “basic principles of New Zealand's legal and constitutional system”, asks whether the legislation complies with fundamental common law principles, whether vested rights have been altered (if so, is that essential, and if so, have compensation mechanisms been included), and whether pre-existing legal situations have been affected, particularly by retroactivity (if so, is that essential, and what mechanisms have been adopted to deal with them).[37]

The guidelines also state “if legislation would implement a taking of property, consideration should be given to whether compensation should be paid to those affected. Where legislation would constitute a taking of property and it is not intended that compensation will be paid, the legislation should make this quite clear.”

While there are these broad constitutional principles in effect, New Zealand does not have a generalised takings regime at the operational level; ie, there is no standard process for assessing or challenging a government action that affects property rights to determine whether the action is essential to the public interest, and if so whether compensation should apply.

There are some specific regimes where takings or compensation issues are identified and regimes put in place, but it is unlikely that these cover all the areas where the benefits of such a regime might exceed the costs. This does not suggest that all government regulation should be so treated, but there may well be scope for wider adoption of such a regime than is the case present.

5.1  Physical takings#

The Public Works Act (PWA) 1981 provides for the compulsory acquisition of land for public works.[38] This reflects very longstanding practice in previous legislation. It is implicitly based on the risk of hold out impeding the completion of essential public works, either physically or by inflating the cost. Compensation is set at current market value. There are also provisions in other legislation for redress for landowners through compensation and or acquisition. These include the Historic Places Act 1980 or the heritage provisions of the Resource Management Act 1991 (RMA).

Public works have previously been defined by reference to an extensive list, based on essentiality, but are currently simply those works which central and local government are empowered to carry out. This approach does not involve any criteria such as hold-out, essentiality or site-specificity. A consequence is that all acquisitions can be seen as carried out under a shadow of compulsion with the accompanying obligations with regard to compensation and possible rights of offer-back when land is no longer required. The acquisition is compensated for at current market value plus certain minor adjustments, such as where a private residence is involved.[39] Access to these powers is currently available to central and local government, and to certain network utility operators.

Private access to such powers can be a contentious issue. As discussed above, however, existing utilities, however, have used these powers to establish their networks and may depend on them for continued access and there may be a public interest in continued operations of such powers to maintain network stability and to allow new entry to the market (Wilkinson, 2001).

5.2  Regulatory takings#

Regulatory takings are likely to be as widespread in New Zealand as anywhere. At the broadest interpretation, every form of regulation of the use of land, labour or capital could be interpreted as a taking through altering vested rights.

Response to biosecurity incursions is an interesting example. The Government takes property when it destroys plants or animals to eradicate or delay the spread of a pest or disease. New Zealand focuses far more on biosecurity than most nations because isolation has protected its territory from such threats and makes continued protection more viable than would otherwise be the case.

By statute the Government compensates for losses suffered as part of eradicating an incursion, but not for any losses suffered prior to detection or as a result of failure to eradicate. Compensation is paid in the first case because it creates an incentive to report infestations. This helps limit total losses to national welfare. It is not equitable, however, for taxpayers to pay for damage they did not cause or have the opportunity to prevent, and the lack of compensation creates an incentive for early detection. Compensation for delays would act as an incentive for the Government to eradicate quickly, but is not equitable to taxpayers who could be liable for a task that was not feasible or which depended on information not obtainable without undue violation of private property rights, and would reduce incentives on property owners to introduce management techniques to minimise their costs.

Taxpayers in turn fund compensation in the first instance due to the public interest in acting rapidly or due to the scope of impact. For narrowly focused incursions, or where funding arrangements are established in advance to allow monitoring, it is possible to charge only those directly benefiting from the response.

Focusing on what lies closest to the accepted boundary of takings for which compensation is payable brings up the RMA. Planning restrictions on land use are the closest non-compensated parallel in New Zealand to the compensated compulsory acquisition of land that occurs under the PWA. The RMA explicitly states that plan provisions are deemed not to have taken or injuriously affected interests in land unless the Act explicitly says so (Ryan, 1998). This is effectively parliament contracting out, through statute, from the application of a takings approach to land use regulation.

Ryan (1998) noted that a compensation provision could be “as bureaucratic and costly to administer as the problems that such legislation would seek to remedy”, but suggested that a “significant” threshold for compensable takings would define tolerable vs intolerable burdens on individuals and encourage regulators to ensure benefits exceeded costs and to spread costs widely.

The earlier discussion in the US context illustrated, however, the complexities of determining when a taking has occurred with regard to this type of property right, where rights continually overlap and community perceptions of the extent of those rights evolve over time. This makes determining and protecting vested rights difficult.

5.3  New property rights#

There are situations in New Zealand where the government has defined new property rights that mean future government action in the area would constitute a compensable taking. A particular example is fisheries, where defined tradable rights have been created.[40] A variation on such arrangements is being considered for aquaculture in New Zealand. This would involve creating involving defined areas within which aquaculture could take place, with the rights to space within those areas to then be subject to tender. This would allow aquaculture to be balanced versus other uses of coastal space, and then ensure that the aquaculture rights go to the most efficient users.

Similar rights have been suggested, and implemented overseas, for controlling pollution as an alternative to common and control (CAC) regulation. In this case, trading of rights allows emission reductions to be made by those able to do so most efficiently, and allows those unable to achieve reductions to buy emission rights from others, to mutual benefit.

In these cases, it could be argued that the government had to restrict its own future behaviour (through creating rights in a form that requires compensation for changes) in order to give property right holders the perception of sufficient security to undertake investment in development and protection of the resource. This is presumably based on an analysis that the benefits of creating those rights, in terms of incentives for management and investment, outweigh the costs.

The primary costs in fiscal terms are (1) future compensation requirements which limit government discretion (waiving the compensation would presumably affect expectations of government treatment of other property rights), and (2) in economic/social/environmental terms any preference which the holders of those defined rights are thereby given over other interested parties with less explicit rights (eg, fisheries quota holders[41] vs recreational fishers or other users of the marine environment).


  • [36]This paper does not address any implications for the use of eminent domain powers of the Crown’s obligations under the Treaty of Waitangi. The issues discussed in this paper arise regardless of Treaty concerns which would require distinct assessment.
  • [37]Legislation Advisory Committee, 2001. The Cabinet Manual notes that Cabinet has endorsed these guidelines, providing essential guidance for everyone involved in preparing and considering legislation, including setting out matters that need to be included in proper drafting instructions. In bidding for legislative slots, Ministers must draw attention to any aspects that have implications for, or may be affected by these guidelines (Cabinet Office).
  • [38]This Act is currently under review (Land Information New Zealand, 2000b and 2001).
  • [39]Compensation for negative side-effects of a public work is only available for those from whom land is taken. Land Information New Zealand (LINZ) publish a booklet on landowners rights when land is required for a public work (Land Information New Zealand, 2000a).
  • [40]This was in response to a situation often described as “the tragedy of the commons” where competition for a scarce common resource results in depletion as no one has a secure property right in future use. Creating exclusive trading rights increases the value of existing rights, allowing compensation for any loss e.g. for conservation reasons or traditional rights, while leaving rights holders still better off than before. Remaining users of the common resource now have an incentive to conserve for the future and operate more efficiently. The rights issued in New Zealand were based on previous fishing records, which were effectively treated as a vested right.
  • [41]Quota reductions must be compensated unless they occur for sustainability reasons.

6  Improving regulatory incentives – alternatives to compensation#

The discussions above have illustrated the difficulties in general of using compensation to improve incentives regarding regulatory action by governments to limit existing property rights, in particular the difficulty of determining which rights are compensable, and the perverse incentives that compensation can create for property owners to invest.

There are also problems regarding the transactions costs of extending compensation eligibility (which is likely to be one factor behind the current position of the compensation boundary) and the fiscal cost of doing so.

The above does not exclude the possibility of addressing specific cases where the benefits of a compensation approach exceeded the costs, but identifying candidates for such treatment is beyond the scope of this paper which focuses on the general situation.

The fallback is to rely on existing structures such as parliamentary decision-making and judicial review, regulatory processes within government and specifically limiting the legal scope for takings to occur. This section discusses these, not mutually exclusive, options.

6.1  Role of parliament and the courts#

It can be argued that specific takings determined by parliament, rather than through administrative means, are more generally accepted due to the representative nature of the parliamentary process and the ability of parliament to provide for compensation. Likewise, as long as courts are seen as impartial, their application and interpretation of the law is generally accepted as not arbitrarily changing, and therefore undermining, property rights structures.[42]

There is a question here of whether there is a public presumption of due process in the courts which allows greater discretion for changes in property rights than would be acceptable in a political process? Under such a presumption, political review of judicial decisions could be considered less acceptable than the converse.

The relative and absolute acceptability of parliamentary and judicial decisions regarding property rights may also be influenced by the existence or not of an overriding constitutional framework, such as applies in the United States where much takings jurisprudence arises. In such a framework, there is likely to be more trust by property rights holders that their interests will not be unreasonably overridden (which may or may not be well founded – the constraints within an unwritten constitution can be highly effective).

The time and expense involved in parliamentary and judicial processes limit, however, their applicability. Inevitably much authority will be delegated and New Zealand has rules for when such delegations should occur and what constraints should be put in place (Legislation Advisory Committee, 2001).

There is no obvious need, or means, to strengthen current arrangements in this area in New Zealand.

6.2  Improving regulatory quality#

The primary regulatory option for constraining takings behaviour is to improve the decision-making processes under which takings are considered. This involves generally adopting more rigorous regimes for improving the quality of new and existing regulation; ie, stricter regulatory governance. The OECD has explored this approach in depth (OECD 1997a, 1997b, 2001).[43] New Zealand has also adopted its own regime based closely on OECD principles and UK and Australian experience.

These regimes include regulatory policies, instruments and institutions. Key policy questions include problem definition, justification for government regulatory action, extent of consultation undertaken, compliance strategy, and an assessment of the costs vs the benefits of the proposal and the feasible alternatives.

The usefulness of cost-benefit analysis in this context has been challenged recently (Wilkinson, 2002), particularly with reference to its dependence on significant and controversial value judgements. Wilkinson also suggests a sequence of tests in two steps. The first step would determine whether a proposal increased individual freedoms of action, contract and exchange, preserved common law causes of action, and preserved existing legal rights and other elements of the rule of law. The second step would check that any violations were essential to pubic well-being, assess whether compensation was required, ensure that any surpluses were distributed proportionately, and eliminate any tying of taxes to permits without explicit parliamentary scrutiny. Such criticisms and suggestions would presumably have to be part of any review of the effectiveness of recent changes to New Zealand’s regulatory quality regime. Those changes followed independent reviews (Ministry of Economic Development, 2001b and 2001c, Tasman Economics, 2001).

In theoretical terms, improving regulatory quality requires re-evaluating both the institutions and instruments involved in making regulation. In this context regulatory institutions include formal governmental bodies at trans-national, national and local level and behavioural constraints such as constitutional principles and the regulatory quality regime.

Regulatory instruments are the tools available for institutions to apply, and include: treaties, primary (statute), secondary (regulation or Order in Council) and tertiary (notices and guidelines) legislation, national instruments for directing local government such as national policy statements or national environmental standards, and regulatory takings. Regulatory impact statements are an instrument used to monitor decisions made in choosing which of the above to apply, and how to do so.

As well as distinguishing institutions from instruments, in order to improve our understanding of how the design of institutions and the choice of regulatory instruments affect incentives on regulators and the quality of the resulting regulatory structures, we need to clearly distinguish upstream and downstream regulatory design.

Upstream regulatory design is about how institutions and instruments create incentives on regulators to design an optimal regulatory structure. Downstream regulatory design is about how that regulatory structure then creates incentives for individuals or firms to act in a manner that achieves a socially efficient outcome. While a discussion of takings involves both perspectives, regulatory quality regimes are an upstream concept.

Regulatory quality regimes in practice tend to involve requiring evidence that consideration has been given to particular factors such as the available alternatives and the wider costs and benefits of the proposal. In New Zealand, this is achieved through requiring preparation of a Regulatory Impact Statement (RIS), or more recently a combined Regulatory Impact Statement/Business Compliance Cost Statement (RIS/BCCS) when primary (statute) or secondary (regulations or Orders-in Council) legislation is proposed. Guidelines on preparing these statements are published (Ministry of Commerce 1996, 1997, 1999 and Ministry of Economic Development, 2001a).

In some cases independent bodies are set up to monitor and enforce regulatory quality regimes, and performance is measured and publicised.[44] New Zealand has moved to increase the transparency of its process by releasing all RISs that contain a BCCS. Regulatory policy quality is also part of the regular assessment of departmental performance in New Zealand by Ministers and Central Agencies.

The key factor in making a regulatory quality regime effective is transparency. If the basis for a regulation and the manner in which it is to be implemented have to be laid out in detail, and those interested in a regulation are able to review and challenge that material, then the outcome is improved incentives on regulators and greater empowerment of the regulated.

Regulatory quality regimes, if effective, act to improve the consideration of the full costs and benefits of proposed regulatory actions, thereby reducing the effect of fiscal illusion on decision-making and increasing the likelihood that only necessary takings will occur. Effectiveness, however, requires robust institutions, which place efficient incentives on the parties involved, and ongoing political commitment.

Avenues for improving the effectiveness of the New Zealand regime could include a stronger oversight role, specifically around monitoring and enforcement, and greater transparency.

6.3  Limiting the use of takings#

A more specific regulatory approach is to target specific cases where takings can occur and impose stricter limits on the ability of government to carry out takings where non-essential. Examples of this would include limiting takings of land to site-specific projects clearly in the public interest, [45] or constraining what types of activities required permission to undertake, or what conditions could be attached to such permission.

Carrying out such an approach would require forming an overall view of what constituted essentiality, identifying the situations to which that definition might be applied, and then working through the implications case by case. A starting point could be to examine situations in New Zealand where adopting a takings perspective from the beginning could have reduced the total costs of achieving a policy objective.

This type of approach would if successful limit the total potential for over-regulation through takings. It does not, however, address whether or not compensation would be appropriate for those activities where takings can still occur (or restrict existing regimes in this area), or restrain the use of other forms of regulation to achieve the same objective (an issue for an overall regulatory quality regime).


  • [42]It has also been suggested that “there is no justification for exempting the judiciary from those property protections that are necessary where other branches of the government are concerned” as “judicial changes in property law raise the same concerns” and this “invites the state to attempt to accomplish through the judiciary what it cannot accomplish through the other branches of government” (Thompson, 1990). This may be more of an issue for a government constrained by a written constitution subject to judicial interpretation than for a parliamentary government in the Westminster tradition.
  • [43]See for more information.
  • [44]The Australian Office of Regulation Review (ORR) is such a body and publishes annual reports.
  • [45]Thereby addressing the holdout problem and providing for network utilities (publicly or privately owned), while avoiding excessive use of compulsion. For non site-specific projects, open-market transactions should be adequate.

7  Conclusion#

This paper has reviewed arguments about the definition of a taking, “the act by which a government assumes or assigns control over all or part of a property right held by a private party”, the protections required against such action, and the role of compensation in such protection. The underlying issue is the crucial issue of security of property rights, in terms of both basic constitutional principles, and the vital role of such rights in providing incentives for investment in and use of the underlying property. Providing security requires clear consistent rules under which regulators can modify those rights in the public interest.

A strict focus on efficiency involves identifying an appropriate standard, and assessing the social costs and benefits of the taking, and of failure to compensate. These assessments carry high information burdens and require judgements on perceptions of a property right and expectations of how rights will change over time. A more case-by-case approach has similar problems in the difficulty of (1) determining when government action ceased being part of an acceptable (constitutionally, legally, or publicly) process for redefinition of property rights,[46] and (2) determining if compensation was appropriate/efficient.

More practically, we can consider the common approach to takings that has evolved here and overseas. First, this accepts that appropriation of physical property (particularly land) must be compensated, unless it is part of generic taxation or overarching changes to rights structures. Second, it defines other property rights as similar enough to physical property or tied to individual ownership, or as so essential for incentive purposes, that equivalent rules should apply eg, tradable fishing quotas. Finally, it leaves other property right regulation to generic non-compensated regulatory processes. This appears to be a “bright line” approach; ie, Government has set an arbitrary rule where easier monitoring and enforcement offset reduced economic efficiency.

Options for New Zealand appear to be either (1) the status quo, or (2) the status quo plus some combination of (A) further tightening of the general regulatory quality regime, (B) more restrictions on takings powers, and (C) extended compensation provisions.

The latter omnibus option provides scope for improvement in protections, while allowing for mitigation of the risks of change – including the changes on incentives of those involved in takings processes. There is likely to be scope for further restricting the ability of government to carry out takings where non-essential or adopting more rigorous regimes for improving the quality of new and existing regulation, without introducing excessive cost or delay in policy making. The latter would complement any restrictions on takings by countering a tendency for regulators to look for other tools to achieve their original objectives, rather than re-evaluating those objectives.

These options 2 (B) and 2 (C) are not, however, quick and easy solutions (nothing in this area is) but require a significant political and administrative commitment. Whether a practical solution also includes a greater role for compensation, ie, option 2(C) requires further consideration, but cannot be completely rejected at this stage.

Whatever change ultimately occurs, if any, needs to recognise the crucial nature of public acceptance of the outcome within a framework of constitutionality and the rule of law. Property rights are heavily dependent on perceptions and highly vulnerable to uncertainty.

Evolution of rights does occur, but the manner in which it does so is crucial to the ongoing legitimacy of the resulting pattern of property rights at any particular point of time. Changes that are too sudden, too large or too unfair (inevitably subjective concepts) can affect that legitimacy.


  • [46]Unusually large or sudden changes, or those which otherwise represent a discontinuity in community perceptions of property rights (as at that point in time) can cross this threshold.


Blume, L, and Daniel L Rubinfeld (1984) “Compensation for Takings: An Economic Analysis. ” California Law Review 72: 569-624.

Blume, L, Daniel L Rubinfeld and P Shapiro, P (1984) “The Taking of Land: When Should Compensation be Paid?” Quarterly Journal of Economics 99:71-92.

Brauneis, R (1996) “The Foundation of Our ‘Regulatory Takings’ Jurisprudence’: The Myth and Meaning of Justice Holmes’s Opinion in Pennsylvania Coal Co. v. Mahon.Yale Law Journal 106:613-702.

Bromley, Daniel W (1993) “Regulatory Takings: Coherent Concept or Logical Contradiction?” Vermont Law Review 17:647-682.

Buchanan, James M (1987) “Public Finance in Democratic Process: Fiscal Institutions and Individual Choice”, The University of North Carolina Press.

Cabinet Office “Cabinet Manual” Wellington, Department of Prime Minister and Cabinet

Davis, GRC (1989) “MagnaCarta” Revised Edition, British Library.

Eposto, Fred G (1996) ”The political economy of taking and just compensation.” Public Choice 89:267-282.

Epstein, Richard A (2000) “Towards a Regulatory Constitution.” Wellington, New Zealand Business Roundtable, April.

Epstein, Richard A (1999) “Property Rights and Takings.” Wellington, New Zealand Business Roundtable, September.

Epstein, Richard A (1985) Takings: Private Property and the Power of Eminent Domain. (Cambridge MA: Harvard University Press).

Fischel, William, A and Perry Shapiro (1988) “Takings, Insurance, and Michelman: Comments on Economic Interpretations of 'Just Compensation' Law.” Journal of Legal Studies 17:269-293.

Fischel, William, A and Perry Shapiro, (1989) “A Constitutional Choice Model of Compensation for Takings.” International Review of Law and Economics, 9:115-128.

Fisher, William W (1988) ”The Significance of Public Perceptions of the Takings Doctrine.” Columbia law Review 88(8):1774-1794.

Grotius, Hugo (1625) “On the Law of War and Peace”, translated from the Latin “De Jure Belli ac Pactis”, by AC Campbell, Batoche Books, 2001.

Guerin, Kevin (2001) “Regulatory Harmonisation – Issues for New Zealand.” Wellington, New Zealand Treasury, Working Paper 01/01, January.>

Guerin, Kevin (2002) “Subsidiarity: Implications for New Zealand” Wellington, New Zealand Treasury, Working Paper 02/03, March.>

Hermalin, Benjamin, E (1995) “An Economic Analysis of Takings.” Journal of Law, Economics and Organization 11(1):64-86.

House of Representatives (1996) “Standing Orders of the House of Representatives”, Wellington, Clerk of The House of Representatives (amended 1999).

Keeney, Jennifer (2002) “Fighting City Hall”, Fortune, May 1.

Land Information New Zealand (2000a) “Landowners' Rights: When the Crown Requires Your Land for a Public Work”, Wellington, Land Information New Zealand.>

Land Information New Zealand (2000b) “Review of the Public Works Act: Issues and Options: Public Discussion Paper ”, Wellington, Land Information New Zealand, December.

Land Information New Zealand (2001) “Review of the Public Works Act: Summary of Submissions ”, Wellington, Land Information New Zealand, August.

Legislation Advisory Committee (2001) “Guidelines on Process and Content of Legislation: 2001 Edition”, Wellington, Ministry of Justice.>

Magnusson, Paul (2002) “The Highest Court You’ve Never Heard Of”, BusinessWeek Asian Edition, April 8.

Miceli, Thomas J and Kathleen Segerson (1998) “Takings” in B Bouckaert and G De Geest eds Encyclopedia of Law and Economics. (Cheltenham: Edward Elgar).>

Miceli, Thomas J and Kathleen Segerson (1994), “Regulatory Takings: When Should Compensation be Paid?” Journal of Legal Studies 23(June): 749-776.

Michelman, Frank I (1967) “Property, Utility and Fairness: Comments on the Ethical Foundations of “Just Compensation” Law.” Harvard Law Review, 80:1165-1258. Reprinted in Bruce A Ackerman ed (1975) Economic Foundations of Property Law. (Boston, Little Brown):100 ff.

Ministry of Commerce (1996) “The Generic Policy Development Process”, Wellington, Ministry of Commerce, December.

Ministry of Commerce (1997) “Code of Good Regulatory Practice”, Wellington, Ministry of Commerce, November.>

Ministry of Commerce (1999) “A Guide to Preparing Regulatory Impact Statements”, Wellington, Ministry of Commerce, March.

Ministry of Economic Development (2001a) “Business Compliance Cost Statements – Guidelines for Departments”, Wellington, Ministry of Economic Development, June.

Ministry of Economic Development (2001b) “Finding the Balance: Maximising Compliance at Minimum Cost. Final report of the Ministerial Panel on Business Compliance Costs”, Wellington, Ministry of Economic Development, July.

Ministry of Economic Development (2001c) “Striking the Balance: Government Response to the Ministerial Panel on Business Compliance Costs”, Wellington, Ministry of Economic Development, December.

Munch, Patricia (1976), “An Economic Analysis of Eminent Domain.” Journal of Political Economy 84(3):473-497.

OECD (1997a) “The OECD Report on Regulatory Reform.” Paris, Organisation for Economic Co-operation and Development.

OECD (1997b) “Regulatory Impact Analysis: Best Practices in OECD Countries.” Paris, Organisation for Economic Co-operation and Development.

OECD (2001) “Regulatory Policies in OECD Countries: From Intervention to Regulatory Governance.” Paris, Organisation for Economic Co-operation and Development, PUMA/REG(2001)10.

Concise Oxford Dictionary: (1995) Ninth Edition (Oxford, Clarendon Press).

Ryan, Kathleen (1998) ”Should the RMA include a Takings Regime.” New Zealand Journal of Environmental Law 2:63-91.

Sax, Joseph L (1964) “Takings and the Police Power.” Yale Law Journal 74(1):36-76.

Sax, Joseph L (1971) “Takings, Private Property and Public Rights.” Yale Law Journal 81(2):149-186.

Stoebuck, William B (1972) ”A General Theory of Eminent Domain.” Washington Law Review 47(4):553-608.

Tasman Economics (2001) “Regulatory Impact Statement Review”, Wellington, Ministry of Economic Development, March.

Thompson, Barton, H (1990) ”Judicial Takings.” Virginia Law Review 76(8):1449-1544.

Wilkinson, Bryce (2001) “Constraining Government Regulation”. Discussion Paper prepared for New Zealand Business Roundtable, Federated Farmers of New Zealand (Inc), Auckland Regional Chamber of Commerce and Industry, Wellington Regional Chamber of Commerce, December.>

Wilkinson, Bryce (2002) “Improving the Quality of Government Regulation”. Paper presented at the Law and Economics session of the New Zealand Association of Economists’ Annual Conference, Wellington, June.

List of Cases#

Lucas v. South Carolina Coastal Council (1992), No. 91-453, 505 U.S. 1003.>

New York Times v. Tasini (2001), No. 00-201 >

Pennsylvania Coal Co. v. Mahon (1922), No. 549,260 US. 393.

Tahoe-Sierra Preservation Council, Inc. v. Tahoe Regional Planning Agency (2002), No. 00-1167.>