|
|
[Home] [Help] [Databases] [WorldLII] [Feedback] |
|
Melbourne University Law Review |
GRAEME HILL[∗]
[This article proposes a new approach to resolving a true conflict between the statutes of different states. It is commonly suggested that this type of conflict should be resolved by applying the statute of the state with the ‘closer connection’ to the dispute. However, states will not legislate unless they have a legitimate interest in the subject matter, so searching for the ‘closer connection’ does not lend itself to any principled answer. Instead, this type of conflict can be resolved by (1) confining a ‘conflict’ between the statutes of different states to the situation when it is impossible to obey or give effect to both statutes; and (2) when there is an actual conflict between the statutes of different states, by giving effect to neither state’s statute to the extent of the inconsistency. This minimalist approach promotes certainty and best gives effect to the equal legislative competence of the states.]
CONTENTS
The Australian Constitution contains a clear rule for resolving an inconsistency between Commonwealth and state statutes — the Commonwealth statute prevails by reason of s 109 — but it does not contain any similar rule for resolving an inconsistency (or ‘conflict’) between the statutes of different states.[1] Of course, as a practical matter, inconsistency between the statutes of different states is much less likely to arise than inconsistency between Commonwealth and state statutes. That is because, for the most part, the legislation of the different states is directed to geographically distinct parts of Australia. Even so, the extraterritorial legislative competence of the states creates the possibility that two or more states might attempt to legislate inconsistently with respect to the same person, thing or event. Consider the following examples:
The High Court has indicated that, in these situations, it would be necessary to resolve any conflict between the statutes of the different states.[5] However, it has not yet been necessary for the Court to determine how this would be done. Several commentators have argued that a conflict between the statutes of different states should be resolved by applying the statute of the state with the closer connection to the subject matter of the dispute.[6] I believe, however, that this ‘closer connection’ test should not be adopted. There are the familiar arguments that the closer connection test is too uncertain, and runs counter to the equal legislative competence of the different states. Of course, those arguments would carry little weight if there were no reasonable alternative method for resolving a conflict between the statutes of different states.
However, I do propose an alternative method, which I call the ‘minimalist’ approach. On that approach:
Under the first proposition, many situations currently thought to give rise to a conflict between the statutes of different states do not involve a true conflict at all. Rather, in the majority of situations, the legislation of different states — even legislation directed at the same person, thing or event — can operate concurrently. However, when there is a true conflict between state statutes, the second proposition is that both statutes should be disregarded to the extent of the conflict. In other words, a true conflict would be resolved by applying any applicable Commonwealth statutes and the common law, together with any other state legislation not in conflict with another state’s statutes. I suggest, perhaps counter-intuitively, that the best way to respect the equal legislative competence of the different states in cases of true conflict is to give effect to the statutes of neither state.
The first question is to determine when there is a true conflict between the statutes of different states. In Port MacDonnell, the High Court indicated that a conflict occurs when there are statutes of two or more states ‘which, by their terms or in their operation, affect the same persons, transactions or relationships’.[7]
In John Pfeiffer Pty Ltd v Rogerson,[8] the joint judgment of Gleeson CJ, Gaudron, McHugh, Gummow and Hayne JJ contrasted a conflict of law and a choice of law. A ‘choice of law’ is said to describe the existence of a possibility of the application of one or other system of law to the facts of the case under consideration. A ‘conflict of law’, on the other hand, is said to describe an inconsistency between laws that leads to the invalidity of one of those laws to the extent of the inconsistency.[9]
Therefore, there seem to be two requirements for a conflict: first, the statutes of two or more states apply in their terms to the same person, thing or event; and second, those statutes are inconsistent with each other.
The first requirement of a conflict (that both states’ statutes apply in their terms to the same person, thing or event) raises two issues: most obviously, whether both statutes apply to the subject matter of the litigation, but also whether both statutes apply in the forum court.
Determining whether both statutes apply to the subject matter of the litigation involves the familiar task of construing apparently general expressions against two statutory presumptions: (1) the presumption against extraterritoriality, and (2) the related presumption that statutes do not apply to matters that, under choice of law rules, would be governed by another law area.[10] The effect of these presumptions is that state statutes usually only apply to persons, things or events ‘in and of’ the state.[11]
In many cases, applying these presumptions will avoid a conflict between the statutes of different states. However, as is well known, the presumption that state legislation only applies to matters ‘in and of’ the state can be overcome by sufficiently clear language. As far as constitutional power is concerned, a state can legislate extraterritorially as long as there is a ‘remote or general’ connection between the state and the subject matter.[12]
Determining whether both statutes apply in the forum court has received less attention. There are some state provisions that can only apply to courts of that state — for example, a right of appeal from a single judge of the Supreme Court of South Australia to the Full Court of South Australia cannot be applied in a New South Wales court.[13] However, even when a state provision can apply in courts in another state, there is still a question as to whether it does apply. On the one hand, there is a presumption that the statutes of one state only apply to courts in that state.[14] On the other hand, the very notion of a conflict between the statutes of different states presupposes that courts in one state are capable of giving legal effect to the statutes of another state; otherwise, a conflict would never arise in practice.
It may be seem like a basic question, but there is vigorous debate on exactly how the statutes of one state are applied or given effect by courts in another state (including courts exercising federal jurisdiction in the other state). Very broadly, there are two approaches to this issue. The traditional analysis for cases in state jurisdiction is that state statutes are applied in the courts of other states through the forum’s choice of law rules.[15] For cases in federal jurisdiction, a substantially similar position is reached by applying ss 79 and 80 of the Judiciary Act 1903 (Cth) (‘Judiciary Act’). This analysis seems to have been accepted in Pfeiffer,[16] although the Constitution may affect the content of these choice of law rules.[17]
On the other hand, there is a longstanding dissenting view that statutes of one state are applicable in courts in other states by force of s 118 of the Constitution. There are two possible versions of this approach: (1) that s 118 itself directly applies the statutes of the other state if they are applicable in their terms (the so-called ‘literal’ or ‘full effect’ view);[18] or (2) that s 118 compels a court to apply the statutes of another state that are applicable in their terms through the forum’s choice of law rules.[19] Although analytically distinct, these different versions of the constitutional approach lead to substantially the same outcomes.[20]
My preferred view is that state statutes apply in courts in other states through the forum’s choice of law rules,[21] and that s 118 of the Constitution merely imposes some constraints on the content of those rules (briefly, those rules must be even-handed as between different law areas). However, the analysis below of how to resolve a conflict between state statutes does not depend on adopting this view. Therefore, I will not continue the debate between the ‘choice of law approach’ and the ‘constitutional approach’ in this part of the article.[22] Instead, I will sketch the apparent consequences of the two approaches and draw out the differences in outcome that follow from adopting one approach rather than the other.
If the statutes of another state apply through the forum’s choice of law rules, then it follows that the only conflict that can arise is between the forum’s ‘substantive’ statute[23] and the forum’s choice of law rule (which picks up the content of the other state’s statute). Two further consequences follow.
First, in practice, a conflict between statutes of different states is translated into a conflict either between a forum statute and the common law (if it is a common law choice of law rule), or between two statutes of the forum (if it is a statutory choice of law rule). It might be thought that on this approach the resolution of the conflict between different states’ statutes depends on the form of the forum’s choice of law rules, rather than anything to do with the substantive statutes of the different states said to be in conflict. However, as I will explain below, that need not be so.[24]
Second, the choice of law approach does not seem to treat some situations as involving a ‘conflict’, even though the statutes of two states are expressed to apply to the same person, thing or event. In particular, a conflict usually arises on this approach because the forum’s choice of law rule is to apply the law of another state. But what if the forum’s choice of law rule is to apply the law of the forum — can the court ignore another state’s statute that is also expressed to apply to that person, thing or event?
Consider Example B (Sid and the TAC — the New South Wales statute applies to New South Wales accidents and the Victorian statute applies to accidents involving Victorian residents). In a New South Wales Court, the New South Wales statute applies in its terms, and the choice of law rule is the law of the place of the wrong (also New South Wales). What happens to the Victorian statute? If the Victorian statute can properly be ignored in this latter situation, then it seems that the choice of venue could resolve an apparent conflict between statutes of different states. Again, as I will explain below, that need not be the case.[25]
By contrast, if the statutes of another state apply by force of s 118 of the Constitution, then a potential conflict will arise between the statutes of different states if they are both expressed to apply to the same person, thing or event, regardless of the forum’s choice of law rules.[26] To this extent, the constitutional approach seems to increase the potential scope for a conflict between the statutes of different states.[27] On this approach, a conflict between statutes of different states will not be affected by the choice of venue.
Two further points about the constitutional approach may be noted. First, this interpretation of s 118 of the Constitution removes the need for choice of law rules in the intranational context. As noted in Pfeiffer, in intranational cases, the only practical significance of choice of law rules is to choose between the statutory law of different states.[28] However, if the constitutional approach is correct, then there is no room for a ‘choice’ because a court in one state is usually compelled by s 118 of the Constitution to apply the statute of another state if that statute is expressed to apply to the person, thing or event in question. Equally, this interpretation of s 118 would seem to make s 51(xxiv) of the Constitution unnecessary, because there seems to be no reason for the Commonwealth to enact legislation providing for the ‘recognition throughout the Commonwealth of the laws [and] the public Acts ... of the States’ if those state laws and public Acts are made directly applicable through s 118.[29]
Second, on the constitutional approach, the only situation in which a court in one state may not be required to apply the statute of another state (even if it is expressed to apply to the person, thing or event) is when there is a true conflict between that statute and the statute of another state. In that situation, the applicable statute will be determined by whatever method the court uses for resolving a conflict between the statutes of different states. Significantly, s 118 of the Constitution itself does not provide any method for resolving that conflict — it simply requires courts to give full faith and credit to the law of every state.
The two different approaches can be summarised as follows. On the choice of law approach, courts in one state will apply the statutes of another state unless the law of the forum provides otherwise (including choice of law rules, both common law and statutory).[30] However, on the constitutional approach, courts in one state will apply the statutes of another state unless the statutes of the forum provide otherwise (including statutory choice of law rules).[31] In the latter situation, there will be a ‘conflict’ that requires resolution.
The key difference between these two approaches, therefore, seems to be the effect given to common law choice of law rules. If this is the key difference, then perhaps the arguments underlying the constitutional approach for applying the statutes of another state only favour a reformulation of the common law, rather than the fashioning of a constitutional implication.[32]
In practice, there has also been a second difference between the choice of law approach and the constitutional approach to applying the statutes of another state. Apart from the different effect given to common law choice of law rules, the two approaches have also been associated with different methods of resolving any conflict that does arise. Specifically, those advocating the choice of law approach have tended to assume that any conflict between the forum law and another state’s statute would be resolved by applying the law of the forum.[33] By contrast, those advocating the constitutional approach have contended for other methods of resolving a conflict — most commonly, a ‘closer connection’ test.[34]
As will become apparent, however, the choice of law approach need not lead to the automatic application of the forum law whenever there is a difference between the statutes of different states. To begin with, it is necessary to determine whether there is any inconsistency between the forum’s substantive statute and the other state’s statute (as picked up by the forum’s choice of law rule).
The second requirement of a ‘conflict’ between the statutes of different states is that the two statutes (both applying to the same person, thing or event) are inconsistent. This second requirement has perhaps not received sufficient attention — in particular, it should not be assumed that statutes of different states are inconsistent merely because they are different.
There is no single test for inconsistency between statutes. For example, the test for inconsistency between statutes of the same legislature (so-called ‘repugnancy’) is subtly different from the test for inconsistency between Commonwealth and state statutes (‘s 109 inconsistency’).[35] Four general propositions can be made about inconsistency between the statutes of different states (‘state–state inconsistency’).
First, state–state inconsistency concerns two statutes of equal status, in the sense that state legislatures derive their authority from the same ultimate source (ss 106–8 of the Constitution) and there is no hierarchy as between the states.[36] This fact favours a relatively narrow test of inconsistency that preserves the operation of both statutes as far as possible.
Second, the fact that the two statutes are drafted by different legislatures means that it is inappropriate to minimise actual contradiction by attempting to ‘read together’ the statutes.[37] If one legislature enacts two apparently contrary statutes but does not make one expressly subject to the other, then the legislature must have considered that the statutes were not inconsistent. There is no reason to suppose, however, that the statutes of two different states were intended to stand together.[38]
Third, no significance should be given to a legislative intention to exclude the operation of other statutory law. Consequently, an inconsistency only arises when it is impossible to obey or give effect to both statutes simultaneously.
Finally, a mere potential for the statutes of different states to operate inconsistently over the same person, thing or event does not give rise to a conflict between those statutes.
The first two propositions should be uncontroversial, but the final two require further elaboration.
As noted above, one general proposition is that the test for state–state inconsistency should not attach any significance to a legislative intention to exclude other statutory law. By comparison, it is appropriate to ask, in the context of s 109 of the Constitution, whether the Commonwealth statute was intended to ‘cover the field’ and so on. This is because s 109 of the Constitution indicates which legislative intention is to be implemented — namely, the Commonwealth’s. In the case of state–state inconsistency, however, there is no basis for deciding which state’s legislative intention should prevail. For example, if both Victoria and New South Wales intended that their motor accidents legislation should operate to the exclusion of other states’ statutes on that topic, these respective legislative intentions would cancel each other out.[39] It may also be noted, by way of analogy, that the test of repugnancy (which, like state–state inconsistency, is concerned with two statutes of equal status) concentrates on the extent to which the two statutes can operate together, rather than any intention to exclude other statutes.[40] Of course, it is still necessary to construe the statutes of both states, and legislative intention will be relevant to this task.[41] The only legislative intention that should be disregarded is the intention to operate to the exclusion of the statutes of other states.
If a legislative intention to exclude the operation of other states’ statutes is disregarded, what types of inconsistency does that preclude? In answering this question, it is helpful to consider the various situations that have been recognised as giving rise to a s 109 inconsistency:[42]
Obviously, the fourth situation (‘covering the field’ inconsistency) depends on a Commonwealth intention to operate to the exclusion of state law. Therefore, it is inappropriate to apply this test when determining whether there is state–state inconsistency.[43] Less obviously, the second and third situations also depend on a Commonwealth intention to operate to the exclusion of state law, even though they are often labelled as examples of ‘direct’ inconsistency. This is because inconsistency will only arise in these situations if the Commonwealth statute confers a positive right. Conversely, there will not be any inconsistency if the Commonwealth statute is intended to operate within the setting of the general law.[44] Therefore, I would argue that there is no constitutional inconsistency between the statutes of different states merely because one state’s statute takes away a right conferred by the statute of another state, or because one state’s statute alters, impairs or detracts from the operation of another state’s statute. That leaves only the situation when one statute requires what another statute forbids — that is, it is impossible to obey or give effect to both statutes simultaneously.
This conclusion — state–state inconsistency is confined to the situation when it is impossible to obey or give effect to both statutes — is consistent with the High Court’s analysis of the extent to which a Commonwealth statute can preserve the concurrent operation of state statutes. In R v Credit Tribunal; Ex parte General Motors Acceptance Corporation Australia, Mason J (with whom Barwick CJ, Gibbs, Stephen, Murphy and Aickin JJ agreed) stated:
It is of course by now well established that a provision in a Commonwealth statute evincing an intention that the statute is not intended to cover the field cannot avoid or eliminate a case of direct inconsistency or collision, of the kind which arises, for example, when Commonwealth and state laws make contradictory provision upon the same topic, making it impossible for both laws to be obeyed.[45]
It might be thought that the use of ‘for example’ in this passage indicates that the Commonwealth cannot legislate to prevent any type of direct inconsistency from arising. However, it is possible that Mason J only meant to leave open the possibility that a general Commonwealth provision preserving the operation of state statutes may sometimes be qualified by other provisions in the Commonwealth Act.[46] In principle, the Commonwealth should be able to legislate so that a s 109 inconsistency only arises when there is what Mason J terms a ‘contradictory provision upon the same topic’ — in other words, when it is impossible to obey or give effect to both the Commonwealth and state statutes simultaneously.[47]
Equally, if it is correct that the test of state–state inconsistency should disregard any intention to exclude the operation of other law, then there will only be an inconsistency between two state statutes when there is an actual contradiction between those statutes. In fact, in the next section, I argue that it may be possible to narrow the meaning of inconsistency even further.
The final proposition about state–state inconsistency is that a mere potential for the statutes of different states to operate inconsistently over the same person, thing or event does not, in itself, create a conflict between those statutes.[48] Instead, I suggest, situations of potential conflict can be dealt with using doctrines analogous to ‘operational’ inconsistency.
In relation to s 109 of the Constitution, ‘operational’ inconsistency describes an inconsistency that results from the particular operation of a Commonwealth and a state statute.[49] This might occur, for example, when a Commonwealth statute confers powers that do not ‘cover the field’ but are intended to be exhaustive once exercised.[50] On this example, the state law is only inoperative once the Commonwealth power is exercised, and even then is inoperative only in relation to the particular person, thing or event over which Commonwealth powers are exercised.[51] The position is similar when Commonwealth and state statutes penalise the same conduct: unless the Commonwealth statute is intended to be exhaustive, there is no s 109 inconsistency between the two statutes and a person can be prosecuted under either statute (but not both).[52]
Applying that reasoning to state–state inconsistency, the fact that the statutes of different states are capable of applying inconsistently to the same subject matter does not in itself create an inconsistency. Rather, the inconsistency only arises if it is sought to apply both statutes to the same person, thing or event.
This argument can be taken one step further. Once legal proceedings relying on one state’s statute are concluded, doctrines such as res judicata or Anshun estoppel[53] (in civil cases) or autrefois convict and autrefois acquit (in criminal cases) should bar any later legal proceedings that seek to rely on the other state’s statute in respect of the same person, thing or event.[54] Of course, res judicata, autrefois acquit and autrefois convict would only be relevant to the extent of the overlap between the statutes of the different states — for example, autrefois acquit and autrefois convict may not provide a complete bar to prosecution if a person is charged later in state B with an offence that has broader elements than an offence in state A with which the person has already been charged.[55] Even so, the operation of res judicata, autrefois acquit and autrefois convict should mean that a conflict is unlikely to arise unless parties seek to rely on both state statutes in the one legal proceeding.
It might be thought that common law principles such as res judicata, autrefois acquit and autrefois convict cannot stand against a legislative intention to apply a state statute to a person, thing or event. There are two responses to this. First, statutes are usually intended to operate against the background of these common law principles. For example, if a person sues at common law in respect of certain subject matter, then any statute that operates on the same subject matter would not ordinarily be construed as permitting the person to sue again in respect of the same matter.[56] So the question is not so much whether there is a clear intention to preserve the operation of common law principles such as res judicata, but whether there is a clear statutory intention to override these principles.
Second, these common law principles may be given additional effect by the requirement that full faith and credit be given throughout the Commonwealth to the judgments of state courts. This derives from a combination of s 118 of the Constitution and Commonwealth legislation enacted under s 51(xxiv) of the Constitution.[57] In Lipohar, Gaudron, Gummow and Hayne JJ indicated that this full faith and credit requirement provides a ‘constitutional footing’ for the doctrines of autrefois acquit and autrefois convict in their application to the judgments of other state courts.[58]
In general, however, there is nothing to prevent a state from supplementing common law principles, such as res judicata, autrefois acquit and autrefois convict, by enacting statutory provisions that deal expressly with an overlap in the operation of the statutes of that state and another state. By analogy, s 4C(2) of the Crimes Act 1914 (Cth) prevents a person from being punished twice for the same act or omission under the statutes of the Commonwealth and a state.[59] There may be other ways of improving the coordination between different states — for example, by introducing a notification requirement for criminal cases that involve an interstate element, similar to the notification of constitutional issues required by s 78B of the Judiciary Act.[60]
The general propositions set out above greatly reduce the situations in which the statutes of different states are found to be in ‘conflict’. Under these propositions, a court would determine whether there is a conflict by asking two related questions:
This general approach can be tested by considering the examples set out earlier.
In Example A, a Queensland company (Canetowed) is prosecuted in New South Wales for a breach of a New South Wales anti-pollution statute, when the polluting activity took place in Queensland and was permitted by the Queensland statute. Assume that both the Queensland and the New South Wales statutes apply to Canetowed’s activity, and also that the Queensland statute can apply in a New South Wales court. Are the Queensland and New South Wales statutes inconsistent?
The earlier analysis suggests that the Queensland and New South Wales statutes will only be inconsistent if Canetowed cannot comply with both statutes simultaneously. Significantly, there will not be any inconsistency simply because the Queensland statute permits an activity that the New South Wales statute prohibits. It is true that, in this situation, the New South Wales statute takes away a right conferred by the Queensland statute, and will ‘alter, impair or detract from’[62] the Queensland statute. However, neither of these types of inconsistency can be taken into account in assessing state–state inconsistency because they depend on a legislative intention to operate to the exclusion of other statutory law.
Therefore, there will only be an inconsistency between the regulatory or penal laws of different states if one state requires (and not merely authorises) an activity that another state prohibits. This type of actual contradiction will probably be unusual, but is not impossible.
I will consider in Part III how this conflict between the Western Australian and South Australian statutes might be resolved.
Example B involved a car accident in New South Wales between a Victorian resident (Melba) and a New South Wales resident (Sid) that was caused by Sid’s negligence. The accident falls within the terms of both New South Wales and Victorian motor accidents statutes, which provide for different compensation schemes. The New South Wales statute expressly preserves the right to common law damages in a negligence action,[64] but limits the amount of damages that can be recovered.[65] The Victorian statute provides an injured person with a statutory right of compensation from the TAC for non-serious injuries.[66] The TAC is given a derivative right to recover that amount from the person who caused the loss suffered by the injured person.[67]
In this case, the TAC sues Sid in the Victorian Supreme Court to recover the money that it paid to Melba. Again, assume that both the Victorian and New South Wales statutes apply in their terms (in particular, that the Victorian statute authorises the TAC’s derivative action against Sid)[68] and that the Victorian Supreme Court is capable of giving effect to the New South Wales statute. It would seem that Sid can ‘comply’ with both the New South Wales and Victorian statutes by not driving negligently. The real issue here therefore seems to be whether a court could give effect to both statutes.
Although the New South Wales and Victorian statutes provide Melba with different claims for compensation, I would argue that these claims are alternatives and are not inconsistent with each other. There is no logical contradiction in a single event giving rise to rights and liabilities arising from different sources.[69]
Most obviously, it is well recognised that there can be overlapping common law and statutory claims. For example, a misrepresentation might give rise to a statutory action under the TPA (or equivalent state or territory legislation),[70] as well as a common law action for negligent misrepresentation or passing off. The fact that the trade practices claim is not made out (say, the misrepresentation was not made in trade and commerce) does not necessarily prevent the common law claim from succeeding.[71] Ultimately, the question is whether the plaintiff can establish any of his or her claims. However, the plaintiff will not recover twice if he or she is successful on both the trade practices and the common law claims. It is only in this sense that the claims are inconsistent.
Equally, there can be overlapping statutory claims. For example, there could be a general statutory prohibition on misleading and deceptive conduct in trade and commerce, and a separate statutory prohibition on misleading or deceptive conduct in issuing company prospectuses. A misrepresentation in a company prospectus could potentially fall within both prohibitions.[72] Although in practice this overlap may be dealt with expressly, the overlap could be resolved even without an express provision. It is highly likely that each statute would be intended to confer ‘once and for all’ compensation for any loss arising from the misrepresentation.[73] In this sense, a court could not give effect to both statutes simultaneously, because neither statute would permit the payment of compensation if the plaintiff had already been compensated for the loss from another source. Consequently, the plaintiff would need to elect between these statutory rights.[74]
Similar reasoning would seem to apply to the New South Wales and Victorian motor accidents statutes. It is likely that a court could not give effect to both statutes simultaneously, in the sense that Melba would not recover under both statutes.[75] However, the mere potential for the Victorian and New South Wales statutes to operate inconsistently in relation to the same motor accident does not create a conflict between those statutes. A conflict will only arise if an attempt is made to apply both statutes to the one accident. Therefore, if Melba applies for compensation under the Victorian statute, the fact that she might have sought compensation under the New South Wales statute is immaterial — she can elect between these inconsistent statutory rights. Once the TAC has recovered the amount of compensation from Sid in the Victorian Supreme Court, that judgment would create a res judicata defence if Melba were then to apply for compensation under the New South Wales statute.[76]
However, the position would have been different if Sid could not have complied with both statutes. If two statutes impose conflicting obligations on a defendant, that inconsistency cannot be resolved by the plaintiff choosing between inconsistent rights.
Admittedly, this analysis does not cover the situation when Melba makes an application under the New South Wales statute before the TAC has obtained judgment under the Victorian statute. Even if the TAC had paid compensation to Melba at that point, Melba’s right to compensation (unlike the TAC’s right to indemnity or damages) is not dependent on anyone being liable to Melba.[77] The TAC’s decision to pay compensation therefore says nothing about Sid’s liability to Melba, which in turn means that the decision does not give rise to a res judicata-type defence.[78] However, even then there are various methods for dealing with a potential overlap between the New South Wales and Victorian statutes:
Therefore, with the law of obligations (such as tort), statutes of different states will usually provide alternative, rather than inconsistent, rights and obligations. A person may recover under either state’s statute, but not both.[81]
However, it will not always be the case that tort statutes confer rights or obligations on the one person. There could be situations where one state’s statute confers an obligation on one person, and another state’s statute confers a converse obligation on another person. Depending on how these obligations are expressed, the obligation under one statute may, in effect, be cancelled out by the obligation under the other statute.
Even if these converse obligations cancel each other out in practice, a court can still give effect to both statutes. There is no legal contradiction in making orders that Canetowed recover under the New South Wales statute and the employees recover under the Queensland statute.[82] Moreover, Canetowed and its employees can comply with both statutes (Canetowed by not instituting the lock out, and the employees by not picketing Canetowed’s factory). In this situation, it may well be the case that each state’s statute alters, impairs or detracts from the other state’s statute. However, as already explained, this does not give rise to a constitutional inconsistency.
There is an important difference between Example E and Example B (involving Sid and Melba). In Example B, I suggested that both statutes were probably intended to confer a ‘once and for all’ compensation so that, once Melba recovered under one statute, the other statute would not be intended to apply. Inconsistency in the operation of these statutes could therefore be avoided by giving effect to this legislative intention. In Example E, it might be assumed that both Queensland and New South Wales intended that the value of their statutory rights (conferred on the employees and employer) should not be reduced by the other party to the industrial dispute receiving compensation from another source. In this situation, inconsistency in the operation of these statutes can only be avoided by not giving effect to this legislative intention. The reason for this difference is that, in Example B, each state has an (implied) intention about when its statute should not apply; by contrast, in Example E each statute has an intention about when another state’s statute should not apply. For the reasons already given, the latter type of legislative intention cannot be given effect when resolving state–state inconsistency.
Tort obligations will be contradictory if doing an act would attract a liability, but not doing that act will also attract a liability. In that situation, the statutes will be inconsistent.
In this situation, there is a contradiction in the obligations imposed on Dr Albury: she can be sued by Melba’s family under the New South Wales statute if she does not provide the blood transfusion, but she can be sued by Melba under the Victorian statute if she does provide the transfusion.[83] A possible means of resolving this conflict is explained in Part III below.
Example C involved a mortgage contract between a Tasmanian resident (Bernie) and the South Australian State Bank, which was expressed to be governed by the law of Tasmania. The South Australian statute permits a court to rewrite an unconscionable contract to which the South Australian State Bank is a party; the Tasmanian statute provides that a contract concerning Tasmanian property is governed by the law of Tasmania. Bernie applies to the South Australian Supreme Court for an order under the South Australian statute rewriting the mortgage contract.
In resolving conflicts between contract statutes, it is likely that there will be more emphasis on whether the courts can give effect to both states’ statutes, rather than whether a person can comply with both statutes.[84] Unlike regulatory and tort statutes — where the emphasis is on the liability of the accused or defendant — contract statutes are concerned with the mutual obligations of both parties to the contract.[85] Consequently, it may be more likely for a contradiction to occur when statutes of different states operate on a single contract. In tort, there is no legal contradiction in a defendant having a defence to one claim but not to another, and therefore being liable to the plaintiff for only one claim. In contract, however, a single contract cannot be both valid and invalid.
In Example C, it may well be the case that a court cannot give effect to the South Australian and Tasmanian statute simultaneously. If Tasmania has not modified the common law of unconscionability, the Tasmanian statute provides Bernie only with the remedies allowed by the common law (as picked up by the Tasmanian statute). By contrast, the South Australian statute permits an unconscionable contract to be rewritten. Even so, the relevant rights provided for by the Tasmanian and South Australian statutes are both conferred on Bernie. He should therefore be able to elect between these inconsistent statutory rights (as with Example B).
In this revised situation, there would seem to be a constitutional inconsistency between the South Australian statute and the Tasmanian statute. The South Australian statute confers a right on Bernie to have an unconscionable contract rewritten, but the Tasmanian statute confers a right on the South Australian State Bank to enforce an ‘unconscionable’ contract. It is not possible to give effect to both of these rights simultaneously. Moreover, unlike Example C, it is not possible to make an election between these inconsistent statutory rights because the rights are held by people with opposing interests. I will consider in Part III below how this conflict between the South Australian and the Tasmanian statutes might be resolved.
Before considering how to resolve a conflict, I should address a possible objection to my proposed test of inconsistency. The test would largely confine inconsistency to situations when it is impossible to obey or give effect to both statutes. It might be objected that this proposed test of inconsistency gives too much opportunity to one state to frustrate the legislative policies of another state. Using the examples given earlier, any permission given by the Queensland statute to engage in an activity in Queensland is effectively negated if a New South Wales statute can penalise that activity. Similarly, a decision by New South Wales that the rights of persons injured in motor accidents in New South Wales should be limited is, at least partly, frustrated if a Victorian injured in New South Wales can recover under a Victorian statute instead. There are, however, several answers to this objection.
First, the objection presupposes that one state will have a ‘primary’ connection with the person, thing or event, usually on the basis of a closer territorial connection.[86] However, a conflict may arise precisely because the different states have a relatively equal connection with the subject matter. For example, if activities in Queensland have an effect in New South Wales, it is not self-evident that one state has a clearly stronger connection than the other. In this situation, it seems appropriate to preserve the operation of both statutes as far as possible by narrowing the area of inconsistency.
Second, the situation when one state clearly does have a stronger connection to the subject matter than another state can be dealt with in other ways. Most obviously, a state cannot legislate with extraterritorial effect unless there is a connection (albeit a remote or general one) with the state.[87] In addition, an implication can be drawn from the federal structure that one state cannot legislate to interfere with another state’s capacity to exercise its governmental functions, which might be called ‘state–state governmental immunity’.[88] These doctrines reduce the scope for conflict between the statutes of different states by limiting the power of one state to legislate on matters that are the responsibility of another state. Therefore, even if it were accepted that there is too much opportunity for one state to encroach upon another state’s area of responsibility, adopting a broader meaning of state–state inconsistency is not the only response. An alternative response would be to develop the doctrines of extraterritoriality and state–state governmental immunity, which could prevent a conflict between the statues of different states arising in the first place.
Finally, there are political means of responding to a state that encroaches on another state’s area of responsibility.[89] In particular, the fact that the major political parties in Australia are organised on a national basis offers a means by which one state can exert political pressure on another state — through the intervention of the federal branch of the political party, if necessary. The weaker the connection between a state and the subject matter, the stronger these political pressures are likely to be.
Part II has proposed a definition of when a conflict arises between the statutes of different states. The proposed definition greatly reduces the area in which the statutes of different states are in ‘conflict’, although there will still be situations when a conflict will arise. In summary:
Once it is determined that there is a conflict between the statutes of different states, the second question is how that conflict should be resolved. Before considering two possible rules for resolving state–state inconsistency, I will briefly explain, and reject, two other possible approaches.
One argument is that state–state inconsistency can be resolved by applying the tests for resolving a repugnancy between statutes of the same legislature.[90] On this approach, the court would apply the more ‘specific’ statute or, as a last resort, apply the statute that is later in time.[91] However, neither of these is appropriate for resolving a state–state inconsistency.[92]
When two statutes are enacted by the same legislature, applying the more specific statute gives effect to the apparent legislative intention that the statutes can coexist. However, there is no reason to suppose that the statutes of different states were intended to coexist.
Alternatively, when two statutes are enacted by the same legislature and cannot otherwise be reconciled, applying the later statute gives effect to parliamentary supremacy, because one parliament cannot bind its successors. As between themselves, however, state legislatures are of equal ‘supremacy’.
A second argument is that a state–state inconsistency could be resolved by the court applying the forum’s statute. There are obviously some practical difficulties associated with this approach. What if neither of the conflicting statutes is enacted by the forum? And what if the forum’s choice of law rule is statutory rather than common law — is the court simply to apply the more specific, or the later, forum statute? That result would create a curious disincentive to enact choice of law rules: the forum’s substantive statute would always trump a common law choice of law rule (that picks up the other state’s statute), but there remains at least a possibility that the forum’s statutory choice of law rule could prevail over the forum’s substantive statute.
Of course, it is true that existing statutory provisions for the transfer and stay of proceedings would ensure that a plaintiff could not determine the outcome simply by choosing the venue.[93] However, by putting such conclusive weight on the venue in determining the applicable law, this approach would preclude a court from responding to other factors that could otherwise influence the choice of venue (such as residence of parties and witnesses, the availability of expedited procedures that benefit both parties, and so on).[94] To this extent, this suggested approach to resolving state–state inconsistency might prevent a court from achieving the fairest outcome between the parties.
There are also conceptual difficulties with attempting to resolve state–state inconsistency by applying the forum statute. The joint judgment in Pfeiffer described a ‘conflict of laws’ as an inconsistency between the statutes of different parts of the federation that leads to the invalidity of one of those statutes to the extent of the inconsistency.[95] The reference to a statute being invalid (presumably in the sense of inoperative) suggests that this conflict is resolved by a constitutional rule, because the statute of one state cannot be invalidated by the statute of another state — or by the common law, for that matter.[96]
Therefore, the better view seems to be that there is an implication drawn from the Constitution itself that the statutes of different states cannot be inconsistent with each other.[97] There are two possible sources of this implication.
First, there is a coherent argument that a constitutional prohibition on
state–state inconsistency derives from territorial limits on state legislative power.[98] In Canada, for example, the provinces have a severely limited power to legislate with extraterritorial effect.[99] This limit largely removes the possibility of a conflict arising between the statutes of different provinces. In Australia, however, it is well established that the states can legislate with extraterritorial effect, including with effect in another state. Moreover, a territorial limit would lead to some practical difficulties, as only one state would ever have power to legislate with respect to any given person, thing or event.
Second, a more widely-accepted argument is that a single legal system cannot contain contradictory commands.[100] Although Australia is a federation with a federal judicial system, the High Court is the ultimate court of appeal for all matters. Partly for that reason, Australia has a single legal system for these purposes. This prohibition on contradictory commands could be seen as an aspect of the rule of law, which has limited constitutional status.[101] Although the rule of law is a notoriously uncertain concept, this particular requirement — an absence of contradiction — seems uncontroversial.
If the Constitution impliedly prohibits a contradiction between the statutes of different states, it is highly doubtful that the constitutional rule for removing that contradiction would depend on the existence of legal proceedings. Consider the position with s 109 inconsistency. Cases such as University of Wollongong v Metwally[102] have emphasised that inconsistency between Commonwealth and state statutes arises directly from the Constitution itself — that is, s 109 operates quite independently of any steps being taken by, or in, a court.[103] Like s 109 inconsistency, the reason that state–state inconsistency must be resolved is to ensure that citizens are not subject to valid but inconsistent commands.[104] It follows that applying the forum’s statute is an unsuitable method for resolving state–state inconsistency, because this approach would not be capable of resolving an inconsistency until legal proceedings had been instituted.[105]
For these reasons, a conflict between the statutes of different states cannot be resolved by applying the tests for resolving repugnancy, or by applying the forum statute. A more difficult question is whether this conflict can be resolved by applying the ‘closer connection’ test.
A common argument is that a conflict between the statutes of different states should be resolved by applying the statute of the state with the ‘closer connection’ to the person, thing or event. There are two different forms of this closer connection test.
The ‘absolute’ form of the test would require that the courts apply the law of the state which has the closest connection to the given person, thing or event. This would be so, even if the law applicable in that state is the common law and another state’s statutes are expressed to apply to the subject matter in dispute.[106] The absolute form therefore seems to assume that for any given person, thing or event, there will be a single state to which this subject matter has the closest connection. The effect of this approach is that only one state will ever have power to legislate with respect to a particular person, thing or event. However, that limit on state legislative power is difficult to reconcile with the states’ well-established power to legislate with extraterritorial effect. Indeed, a power to legislate extraterritorially seems, inevitably, to create the possibility that more than one state will have power to legislate with respect to the same subject matter.[107] Therefore, in the following discussion, I focus on the ‘comparative’ form of the closer connection test.
The comparative form of the test requires that when two states legislate with respect to the same person, thing or event, the courts should apply the statute of the state that has the closer connection to the subject matter. Unlike the absolute form of the test, the comparative closer connection test only has regard to statutory law — the courts should give effect to a state statute, even if there is another state with a potentially closer connection that has not legislated on the subject matter.[108]
It is convenient to start with the main arguments in favour of a closer connection test. In summary, the arguments are that the closer connection test:
First, the closer connection test receives some support from High Court authority. There are remarks in Port MacDonnell that suggest that a conflict between the statutes of different states should be resolved by determining which state has a closer connection to the particular person, thing or event. After stating that there was no s 109 inconsistency in that case, the Court continued:
A problem of greater difficulty would have arisen if the fishery defined [pursuant to the relevant South Australian Act] had a real connection with two States, each of which enacted a law for the management of the fishery. If the [South Australian instrument defining the fishery] had been construed as extending to waters on the Victorian side of the line of equidistance, there would obviously have been grounds for arguing that the Victorian nexus with activities in these waters was as strong as or stronger than the South Australian nexus. ... In the present case, [however,] ... there is no real question of any relevant inconsistency between the law of South Australia and the law of another State.[109]
It is clear from this passage, however, that the Court was not offering a concluded view on how a hypothetical conflict between Victorian and South Australian statutes should be resolved. In any event, to the extent that the High Court was using territoriality as the main criterion for resolving state–state inconsistency, this argument is considered separately below.
A second argument is that if one state’s statute is to be given priority over another in cases of conflict, it is appropriate to assess the relative strength of each state’s connection with the relevant person, thing or event.[110] This would indeed be a powerful argument if one state’s statute were to be given priority. However, if the alternative is giving effect to neither state statute to the extent of the inconsistency, this second argument does not take the matter much further.
A third argument in favour of a closer connection test is that the test leads to the same result regardless of the choice of venue (as opposed to, say, applying the law of the forum). This argument should not be confused with the more general argument that conduct in Australia must lead to a single predictable outcome in Australian courts, regardless of where those proceedings are brought.[111] It may be accepted that uniformity of outcome is a desirable objective in formulating common law choice of law rules. As Kirby J in Pfeiffer explained, if a choice of law rule permitted different results, depending on venue, ‘the law would no longer provide a certain and predictable norm, neutrally applied as between the parties. Instead, it would afford a variable rule which particular parties could manipulate to their own advantage.’[112]
However, the later case of Blunden v Commonwealth[113] strongly suggests that uniformity of outcome may be desirable, but it is not a constitutional requirement, even for cases in federal jurisdiction. Moreover, ‘forum-shopping’ can be addressed by regulating the choice of venue, rather than the choice of law.[114]
Instead, the reason that the rule for resolving state–state inconsistency must be ‘venue-neutral’ is because that rule operates prior to any legal proceedings being instituted. In any event, the minimalist approach proposed below — giving effect to neither state statute to the extent of the inconsistency — is ‘venue-neutral’ in this sense. Therefore, this third argument is not a reason to favour the closer connection test over the minimalist approach proposed below.
Perhaps most significantly, the closer connection test reflects the ‘predominant concern’[115] that each state is said to have with matters within its geographical area. Therefore, on this argument, the test for resolving state–state inconsistency should give priority to a statute based on a predominant connection with the subject matter in question, rather than legislation based on a remote or general connection.
It may be accepted that the federal structure is framed on the assumption that the states will generally legislate only with respect to their own geographical areas. That assumption is given effect through the requirement that a state cannot legislate with extraterritorial effect unless there is a (remote or general) connection between the subject matter and the state. For the following reasons, however, I do not think that territoriality should be given additional weight by using it as the primary criterion for resolving a conflict between the statutes of different states.
First, in some situations, it will be difficult to say that one state has a closer territorial connection than another state. In Example A, the Queensland statute regulates activities that occur in Queensland and the New South Wales statute penalises activities that have an effect in New South Wales.[116] Moreover, other matters, such as interstate communications or transactions, do not have an obvious location.[117] Asking which state has a closer connection to the relevant subject matter in either of these situations does not lend itself to any particular answer.
Secondly, even if one state clearly has a closer territorial connection, it is not self-evident that a closer territorial connection should always outweigh other connections, such as a connection based on residence. For example, although Pfeiffer held that applying the law of the place of the wrong strikes the appropriate balance between the interests of the various states,[118] some states have since legislated so that cross-border workplace accidents are governed by the law of the state where the employee is insured, rather than the law of the state where the accident occurred.[119] Moreover, there may be two important, but competing, government interests underlying each statute. In Example D,[120] Western Australia has an interest in court proceedings that take place in Western Australia, but South Australia has an interest in social work activities in South Australia. Similarly, in Example C,[121] Tasmania has an interest in real property in Tasmania, but South Australia has an interest in the activities of its state bank.
In these situations, as Callinan J observed in Mobil Oil, the different states have ‘legislate[d] in respect of matters with which each has a legitimate connection’.[122] The reference to a ‘legitimate’ connection emphasises that a state–state inconsistency cannot arise unless both state statutes have a sufficient territorial connection with the enacting state to be valid, and the statutes are consistent with other constitutional limitations, such as state–state governmental immunity.[123]
Thirdly, as noted already, the constitutional prohibition on state–state inconsistency seems to be derived from the rule of law rather than from territoriality. Therefore, although it would be appropriate (and indeed necessary) to give territoriality primary weight if the prohibition on state–state inconsistency derived from territorial limits on state power, territoriality would be invoked here only as a ‘tie breaker’ to resolve state–state inconsistency, rather than as a constitutional imperative.
It is true that a closer connection test need not be limited only to territorial considerations. However, to the extent that ‘closer connection’ includes factors other than a territorial connection, it can no longer rely on the ‘predominant concern’ that each state is said to have in matters within its geographical area. Indeed, the fact that the closer connection test requires the weighing of a large number of different factors might be a criticism of that test.
The main arguments against the closer connection test can be summarised as follows:
It is useful to consider together the first two of these arguments: the lack of certainty inherent in the closer connection test, and the limits the test places on a state’s ability to use connecting factors that are adapted to particular situations. This is because there tends to be a trade-off between maximising certainty and maximising legislative options. The two arguments can be summarised as follows:
There is some force to each of these criticisms. Consider, for example, two tests that have been proposed for deciding which state has the closer connection: (1) the test of ‘predominant territorial nexus’, balanced against what is ‘fair and just’;[124] and (2) ‘the strength of the governmental interest in regulating a matter connected to the governance of the polity’s territory’, as assessed on a case-by-case basis.[125] Obviously, it is very difficult to predict the outcome of such open-ended tests. However, if, as seems likely, the courts attempt to increase certainty by adopting broad rules, those rules may not be suitable for a particular case. For example, imagine that the courts adopted a rule that torts were most closely connected with the state where the tort occurred. This rule would not necessarily be suitable for all intranational torts — by analogy, lower courts in Canada have held that applying the lex loci delicti[126] is not suitable for all inter-province torts, such as a misrepresentation in a company prospectus.[127] However, as it is a constitutional rule on this hypothesis, the lex loci delicti rule could not be amended. To the extent that courts move away from general rules to respond to particular cases, this ad hoc response will decrease the certainty of the closer connection test.[128]
Nonetheless, while these criticisms of the closer connection test have some force, they are unlikely to be decisive in themselves. There is uncertainty in all law, including constitutional law.[129] Equally, limiting legislative options to some degree is an inevitable result of having a constitutional rule to resolve state–state inconsistency. Even so, certainty and maintaining legislative options are both important objectives.[130] Therefore, other things being equal, the fact that a test increases certainty and legislative options would be a reason, perhaps a strong reason, to prefer that test over others.
A related argument is that, in practice, the closer connection test is difficult to distinguish from a bare assessment of the relative merits of the state legislation. Arguably, assessing which state has a closer connection to a given person, thing or event is not an appropriate task for the courts.[131] A conflict between the statutes of different states will ordinarily arise when the different states have based their legislation on different, but legitimate, connecting factors. For example, one state might legislate on the basis of where events occur, while another state might legislate on the basis of the residence of the people involved. A ‘closer connection’ test requires the courts to decide which of these connecting factors should be given priority. This requirement seems to skirt uncomfortably close to asking the courts to decide which of the competing legislative policies is more important. Clearly, this is a decision that the courts are ill-equipped to make and, in any event, should not be making.[132]
One possible response to this argument is that the ‘closer connection’ test is not significantly different from matters already determined by the courts, such as the test for the validity of extraterritorial state legislation, and the common law choice of law rules.
However, there seems to be only a superficial similarity between the closer connection test and the test for extraterritorial state legislation. Asking whether a state law has a (remote or general) connection with the state does not require the courts to assess the relative strength of different connections with a state, and to give priority to different, legitimate connecting factors.
On the other hand, there is arguably a stronger similarity between the closer connection test and common law choice of law rules. Like the closer connection test, common law choice of law rules involve an assessment of which state (or ‘law area’) is more closely connected with a person, thing or event, having regard to various different connecting factors. Moreover, common law choice of law rules are also framed by reference to general categories, such as tort, contract and so on. There is, however, an important difference of degree. Although the closer connection test will probably lead to the courts developing general rules, it also seems likely that these rules would be subject to qualifications that enable the courts to respond to a particular case. For example, one test of closer connection mentioned earlier is ‘predominant territorial nexus’, subject to what is ‘fair and just’.[133] Such ad hoc qualifications could get dangerously close to a bare assessment of the respective merits of the different state statutes.
On balance, this argument probably does not undermine the closer connection test altogether. Rather, the concern that courts should not judge the relative merits of particular legislation may only be a reason to favour a relatively rule-based application of the closer connection test, rather than an ad hoc approach.
A final point is that the closer connection test runs counter to the equal legislative competence of the states because it requires that priority be given to one state’s statute over another state’s statute. Of course, the equal legislative competence of the states could not permit a true conflict between the statutes of different states to stand.[134] However, if the conflict could be resolved without giving priority to one state’s statute, then the states’ equal legislative competence would be a reason to prefer that method. In addition, a constitutional rule that assigns priority to one state’s statute over another state’s statute reduces the incentives for the states to resolve an inconsistency between their statutes through cooperative political means.[135]
As might be expected, these arguments for and against the closer connection test do not strongly indicate whether that test should be used to resolve
state–state inconsistency. Instead, the real issue seems to be whether the closer connection test is better than the available alternatives. The following discussion sets out an alternative approach to resolving state–state inconsistency (giving effect to neither state’s statute to the extent of the inconsistency) and then compares that approach to the closer connection test.
As already noted, there has been a tendency to assume that a conflict between the statutes of different states can only be resolved by giving priority to one state’s statute. For example, the joint judgment in Pfeiffer defined a ‘conflict of law’ as an inconsistency between laws that ‘lead[s], to the extent of the inconsistency, to the invalidity of one law.’[136] A similar assumption underlies the following description of the approaches available under the ‘full faith and credit’ clause in art IV, § 1 of the United States Constitution: ‘To simultaneously apply the conflicting law of two states is impossible; to require each state to apply the law of the other is absurd; and to let each state apply its own law repeals the Clause.’[137]
The following discussion challenges that assumption and suggests that a conflict between the statutes of different states can be resolved by not giving effect to either state’s statute to the extent of the inconsistency.
One obvious objection to this suggested approach is that giving effect to neither state’s statute would be unworkable and would lead to chaos. This would not be so. The minimalist approach would operate as follows:
Of course, this general outline raises some questions, such as the meaning of ‘liability’ in this context, and what is meant by ‘contradictory’ rights. Those questions will be addressed in considering how the proposed minimalist approach would apply in different situations.
Concerns about legal chaos carry particular weight in the context of regulatory and penal laws. However, for the reasons that follow, the suggested approach should not lead to chaos.
First, and most importantly, a ‘conflict’ will only arise if the statutes of different states are inconsistent. In the context of regulatory and penal statutes, there is only an inconsistency if one state requires (and not merely permits) what another state prohibits.[138] Two consequences follow. To begin with, this type of actual contradiction between the regulatory statutes of different states should be very rare. The fact that neither state’s statute will be given effect in isolated cases should not undermine the general administration of either statute. Moreover, the courts will only be refusing to give effect to both states’ statutes when a person is subject to contradictory and apparently valid commands. Being subject to different contradictory and apparently valid commands hardly encourages compliance with the law. Even if the person knew in that situation that only one of those statutes would be given effect by the courts, it may be difficult for the person to predict which statute that would be.
Second, the conflict between the statutes of different states would only invalidate each statute to the extent of the inconsistency. Significantly, the inconsistency would only arise to the extent that one statute required what another statute prohibited. Therefore, a person who ignored both the statutory requirement and the statutory prohibition would not come within the area of the inconsistency and could be prosecuted under either statute. In effect, a person would need to comply with one statute to avoid prosecution (although it would not matter which one). Two further points may be noted:
Returning to the examples given earlier, this approach to resolving state–state inconsistency would not lead to the invalidity of Queensland and New South Wales anti-pollution statutes in Example A. There is no inconsistency between those statutes, and therefore no conflict to resolve.
In Example D, there does seem to be a constitutional inconsistency between the Western Australian and South Australian statutes. The Western Australian statute requires Adelaide to provide certain information to the court, but the South Australian statute prohibits Adelaide from disclosing that information to a court. On the minimalist approach, each state’s statute would be inoperative to the extent that it purported to penalise conduct that was required by the other state’s statute. Therefore, Adelaide could choose either to comply with the Western Australian statute (and provide the information) or to comply with the South Australian statute (and refuse to provide the information). As long as she complied with one state’s statute, she could not be prosecuted for breach of the other state’s statute.
It might be objected that this approach would require a state court to apply a hybrid of Western Australian and South Australian statutory law.[140] However, that is not the case, because Adelaide has a constitutional — not statutory — defence to prosecution. Imagine that she complied with the South Australian statute, and she is prosecuted in Western Australian for failing to disclose the information. Adelaide’s defence is that the Western Australian statute cannot validly penalise her conduct, because of the implied constitutional prohibition on contradictory state statutes. The South Australian statute, in itself, does not provide a defence to prosecution under the Western Australian statute.
A similar analysis would apply to a conflict between state tort statutes. Once again, the situations when there is an actual conflict between the tort statutes of different states will be rare indeed. Tort statutes are not inconsistent merely because they impose different, alternative, obligations on one person and that person is able to comply with both. Moreover, tort statutes are not inconsistent merely because they impose converse obligations on different people, provided that the court can give effect to both statutes. However, there will be a conflict between the tort statutes of different states if the statutes impose contradictory obligations on a person such that doing an act attracts liability under one statute, but not doing that act attracts liability under the other.[141]
The minimalist approach would read each state statute down to the extent of the inconsistency. Accordingly, when the tort statutes of different states impose conflicting obligations, each state’s statute would be inoperative to the extent that it attempted to impose a liability on conduct that was required to avoid a liability under the other state’s statute.[142]
A difficult question arises here about what sort of liability would be sufficient to give rise to a state–state inconsistency. For the following reasons, not every non-trivial adverse consequence should be treated as a ‘liability’ for these purposes. A ‘liability’ would clearly include a liability for an offence (whether civil or criminal) and would probably include tort-like liability. Consequently, a state–state inconsistency could conceivably arise between a statute that imposes liability for a criminal or civil offence and another statute that imposes tortious liability. To require a person to choose between committing a tort or a crime does not seem like a fair choice.[143]
The difficult question is whether ‘liability’ should include other forms of liability, such as contractual liability. In principle, there does not seem to be any reason to treat a contractual liability to pay compensation differently from a tortious liability.[144] However, what about a non-monetary penalty such as ‘shaming’, or loss of a statutory benefit? The issue cannot be whether an adverse consequence has the same economic effect as a liability (such as a fine), because otherwise a taxation liability could give rise to a state–state inconsistency.
It therefore seems that a ‘liability’ would definitely include a fine and an obligation to compensate a person for loss, and that other adverse consequences would need to be considered closely. Ultimately, this issue seems to turn on a value judgement as to whether it would be fundamentally unfair to place a person in the position of incurring one adverse consequence rather than another. This appeal to fairness may explain why a taxation liability should not be treated as a liability for these purposes.[145] It seems unfair if we cannot structure our affairs to avoid committing an offence, tort or breach of a contract. Conversely, there does not seem to be any comparable unfairness if we cannot structure our affairs to avoid paying a tax.[146]
Returning to the examples considered earlier, this approach to resolving
state–state inconsistency would not mean in Example B that the New South Wales and Victorian motor accident statutes were both inoperative. As already noted, there was no true conflict between the statutes because Melba could choose between conflicting rights.
In Example F, the New South Wales and Victorian statutes do seem to impose conflicting obligations. Dr Albury could be sued by Melba’s family under a New South Wales statute if she did not provide a blood transfusion to Melba and Melba died (even when Melba had not consented to the transfusion), but she could also be sued by Melba under the Victorian statute if she did provide the transfusion without Melba’s consent. As an aside, it may be noted that a conflict may involve the defendant owing contradictory obligations to different people — in this example, to both Melba and Melba’s family.
On the minimalist approach, the conflict between these various obligations would be resolved by making both statutes inoperative to the extent of the inconsistency. The New South Wales statute would not impose on a doctor a duty to provide treatment when providing that treatment would give rise to a liability under the Victorian statute, and the Victorian statute would not impose liability on a doctor for providing treatment without consent when failing to provide treatment would give rise to a liability under the New South Wales statute. Dr Albury would still need to comply with one of either the New South Wales or Victorian statutes in order to avoid liability under the other statute. So, to rely on the New South Wales statute, Dr Albury would need to show that the blood transfusion was necessary to save Melba’s life (if treatment was given without consent). To rely on the Victorian statute, Dr Albury would need to show that Melba refused consent, if the withheld blood transfusion could have saved Melba’s life.
Again, this approach would not require a court to apply a hybrid of New South Wales and Victorian statutory law. Imagine that Melba’s family sues Dr Albury under the New South Wales statute for failing to provide a transfusion. The New South Wales statute cannot validly impose a liability on Dr Albury, because in that particular purported operation the New South Wales statute is contrary to an implied constitutional prohibition against contradictory state statutes. The action brought by Melba’s family under the New South Wales statute therefore fails.
The main issue with contract statutes will usually be whether the court can give effect to both statutes. In situations when the courts cannot give effect to the conflicting contractual rights conferred by the statutes of different states, each state statute would be inoperative to the extent that it attempted to confer a right that was contradictory to a right conferred by the other state’s statute.
It is necessary to explain here what is meant by ‘contradictory’ rights. As noted earlier, rights are not necessarily contradictory merely because one right alters, impairs or detracts from the other right. Moreover, if rights are conferred on one particular person, any inconsistency can be resolved by an election between those rights. The question therefore is whether, if both parties were to rely on these different rights in one instance, it would be possible for the court to make orders giving effect to both rights. By analogy, in Example E, it was possible for the court to order under the Queensland statute that the employer pay its employees lost wages, but also to order under the New South Wales statute that the employees pay the employer lost profits. There was therefore no constitutional inconsistency between the rights conferred under both statutes, even though in practice the rights might cancel each other out.
However, there need not be a precise correlation between rights for those rights to be ‘contradictory’ in this context. For example, one state’s statute might permit a court to rewrite a ‘harsh or unreasonable’ contract, and another state’s statute might permit a person to enforce an ‘unconscionable’ contract. Although there is no precise correlation between a contract being harsh and unreasonable and being unconscionable, these different concepts overlap to a significant degree. Therefore, to the extent that the harsh or unreasonable test overlapped with unconscionability, but no further, there would be an inconsistency between these statutes. However, it must be accepted that the question of whether different rights overlap will sometimes involve difficult judgements, including the appropriate level of generality with which to describe the two rights.[147]
These points can be illustrated by Example G. A South Australian statute permits a court to rewrite an unconscionable contract to which the South Australian State Bank is a party, but the Tasmanian statute provides that unconscionability is not a ground for invalidating or otherwise affecting the validity of a contract concerning Tasmanian land. Each state’s statute would be inoperative to the extent that it conferred a right that was contradictory to a right conferred by the other state’s statute. The right to have an unconscionable contract rewritten and the right to enforce an unconscionable contract are contradictory in the relevant sense because a court could not make orders that gave effect to both rights. In effect, this aspect of the dispute would be governed by the common law of unconscionability.
It may seem odd that, even though South Australia and Tasmania have both legislated to change the common law, it is ultimately the common law that is applied to the dispute. That result does not depend, however, on the common law being some sort of ‘pre-political’ baseline;[148] rather, it is a matter of necessity. If a court cannot give effect to both states’ modifications to the common law — and assuming that there are no relevant Commonwealth statutes — then the court has no resource other than the common law with which to determine the dispute.
The preceding statement raises a further question: once an inconsistency between the statutes of different states is removed, which law then governs the dispute? In general terms, the minimalist approach does not affect whether a state statute is applied, or which state statute is applied — it affects only the content of any state statute that is selected. Accordingly, the law governing the dispute is determined in the usual way: by applying any relevant Commonwealth statutes, then any relevant state statutes — including statutes of another state applied by the forum’s choice of law rules — and finally the common law. Two specific points require further elaboration.
First, a state–state inconsistency will only render the statutes inoperative to the extent of the inconsistency. It will ordinarily only be the particular provisions creating the contradictory rights or obligations that will be inoperative, and the remainder of the statute will continue to apply.[149] In some cases, however, the ‘invalidity’ of one provision may lead to the consequent invalidity of another provision that is legally dependent on the invalid provision. By analogy, an inconsistency under s 109 of the Constitution has the following effect:
every part of a completely interdependent and inseparable legislative provision must fall within ‘the extent of the inconsistency.’ No doubt s 109 means a separation to be made of the inconsistent parts from the consistent parts of a State law. But it does not intend the separation to be made where division is only possible at the cost of producing provisions which the State parliament never intended to enact. The burden of establishing interdependence in such a case is necessarily upon those who assert it in view of the words of s 109, and perhaps it is not a light one.[150]
Similarly, s 79 of the Judiciary Act will not pick up ‘some but not all of [a state statute], if to do so would give an altered meaning to the severed part’ of the statute.[151]
Second, the minimalist approach will affect the content of a state statute, even when that statute is picked up by the forum’s choice of law rules. Assume for the moment that the statutes of one state are applicable in the courts of another state through the forum’s choice of law rules.[152] As noted earlier, a constitutional rule for resolving state–state inconsistency does not depend on the existence of legal proceedings. Instead, the constitutional implication underlying the minimalist approach removes the inconsistency prior to a state statute being picked up by the choice of law rules of another state. By way of analogy, when there is a s 109 inconsistency between a Commonwealth and a state statute, the state statute is not picked up and applied to the Commonwealth through s 64 of the Judiciary Act.[153] Consequently, when there is a state–state inconsistency, the choice of law rule would pick up the statute as ‘read down’ to the extent of the state–state inconsistency. The minimalist approach would not, however, affect which statute was selected by the forum’s choice of law rule.
The significance of these points can be illustrated by returning to Example F. It will be recalled that Dr Albury decides not to give Melba the blood transfusion, because Melba had refused consent, and Melba dies. Dr Albury’s conduct avoids an apparent liability to Melba under the Victorian statute, but creates an apparent liability to Melba’s family under the New South Wales statute.
Imagine now that Melba’s family live in Queensland, and wish to sue Dr Albury under the New South Wales statute, either in New South Wales, Victoria or Queensland. (Assume that the New South Wales statute confers a right on relatives who live outside New South Wales.) None of these states have modified the common law choice of law rule that intranational torts are determined by the place of the wrong.
Under the minimalist approach, both the New South Wales and Victorian statutes would be read down to the extent of the state–state inconsistency. Relevantly, the New South Wales statute would be inoperative to the extent that it purported to impose liability on a doctor who did not provide the transfusions in order to avoid incurring a liability under another state’s statute. Assuming that the court in each state adopted the ‘choice of law’ approach to applying the statutes of another state, the outcome of the lawsuit by Melba’s family in the various venues would be as follows:
Therefore, it can be seen that the operation of the minimalist approach should not be affected by the choice of venue. Moreover, as the minimalist approach — unlike the closer connection test — does not require the forum court to apply any particular statute, there is no tension or contradiction between using this method of resolving state–state inconsistency and using the ‘choice of law’ approach to apply the statutes of other states.
Assuming that the minimalist approach is workable, the next question is whether it is a better method of resolving state–state inconsistency than the closer connection test. To a large extent, the advantages and disadvantages of the minimalist approach are the reverse of the advantages and disadvantages of the closer connection test discussed earlier. On one hand, the minimalist approach seems to be relatively certain[154] and avoids placing the courts in a position where they have to decide that one state’s statute should be given priority over another state’s statute. On the other hand, the minimalist approach runs counter to suggestions from the High Court that each state has a predominant concern with respect to matters within its geographical area. I have already argued that considerations of territoriality need not be given primary weight in resolving a state–state inconsistency.
Another argument against the closer connection test is that it would limit the options for parliaments to use connecting factors that are suitable for particular situations. Of course, limiting legislative options to some extent is the inevitable result of using a constitutional rule to resolve state–state inconsistency. However, as a practical matter the minimalist approach will probably reduce legislative options to a lesser degree than the closer connection test.
First, the minimalist approach is combined with a much narrower test of
state–state inconsistency than is usually proposed for the closer connection test.[155] This narrow test of inconsistency preserves the operation of competing state statutes as far as possible, and to this extent permits the use of different connecting factors.
Second, the minimalist approach — on my preferred analysis — is combined with an approach to applying the statutes of another state that allows states to enact choice of law rules. The closer connection test, by contrast, is often combined with a constitutional approach to applying the statutes of other states that leaves no room for intranational choice of law rules.[156]
Admittedly, these practical differences do not result from anything logically inherent in the closer connection test. If the closer connection test had a similarly narrow test of state–state inconsistency and was combined with the choice of law approach to applying statutes of another state, then there may be little practical difference between that test and the minimalist approach in the limits they place on legislative options.[157]
Considering the practical operation of the closer connection test and the minimalist approach, it would seem that the different methods for resolving
state–state inconsistency function better in different situations.
The closer connection test works particularly well when one state clearly has a stronger interest, because there is a strong intuition that the law of that state should prevail. That intuition might hold even when it is technically possible to