I. Introduction

The nature and scope of executive power tend to be elusive in any democratic constitutional system. Executive power typically has a residual character, defined by reference to what legislative and judicial power are not. The actions that can lawfully be undertaken in the exercise of executive power may be diverse and open-ended and defy neat categorisation. The executive branch itself is sprawling and amorphous, offering a catch-all for public bodies that do not neatly fit elsewhere under a tripartite separation of powers. Most executive decision-making takes place behind closed doors, often in conditions of confidentiality, in contrast to the relative openness of at least the formal processes of legislatures and courts.

In polities that are constitutionally organised to give effect to the federal idea the challenge of understanding executive power is greater still. In federal-type systems, executive power necessarily is shared between at least two orders of government, each with their own executive organs. Whatever complexities are inherent in the conception of executive power thus are augmented by its interdependence with federal arrangements. As neither the conception of executive power nor the form through which the federal idea is given effect can be assumed to be standard across polities, an appreciation of the nature and scope of executive power in federal-type systems depends on variations along at least two axes. Each of these axes offers a range of possibilities that have implications for the scope and operation of executive power, even in isolation from each other. The point at which they intersect in a particular federation is likely to produce results that are distinctive as well.

One axis concerns the design of the federation itself. Along this axis, there is a pronounced distinction between dualist federations, in which each sphere of government is expected to execute its own legislation and integrated federations, in which a significant proportion of central legislation is executed by other spheres. The United States and Germany can be regarded, respectively, as exemplars of these two approaches, with all other federal-type systems ranged somewhere between the two, each with variations on the theme of their own.

The second axis along which the nature and scope of executive power reflects the wider constitutional setting of which the provisions for federalism are part. This setting includes, critically, the relationship between the executive and the legislature within each of the orders of government and, potentially, between the executive and the courts as well. For this reason, in this essay, I characterise variations along this axis in terms of the separation of powers. In examining executive power from this perspective, I focus not only on the execution of legislation but also on the power of the executive to act without legislative authority. In addition, however, there may be other aspects of the constitutional context that affect the scope of executive power in federal-type systems. These include underlying ideas about the relative scope of legislative and executive power that the Constitution assumes and may reinforce and any other, specific, constitutional provisions that have interpretive value in understanding executive power.

In what follows, I examine variations along each of these axes to identify the similarities and differences to which they give rise, individually and collectively, and to demonstrate their relevance to the nature and scope of executive power. In a final substantive part I use Australia as an extended case study, to bring depth to the treatment of the subject and to illustrate the practical relevance of the variations along both axes in the experience of one federated state. The scope of federal executive power is a live question in Australia, the emergent answers to which are distinctive, from a comparative point of view. In closing, I draw some brief conclusions from this study for the purposes of comparative federalism.

I am conscious that in developing the arguments in this essay I am influenced by an Australian frame of reference, which may well not reflect approaches to thinking about executive power in federal settings elsewhere. Achieving a satisfactory level of objectivity is a challenge in any comparative exercise and is particularly difficult in the present context, where so many variables stem from history, theory and practice and shared terms mask what may be underlying difference. Even to understand that frameworks of reference may differ, however, is to make some progress in comparative terms. I hope that this contribution prompts others to engage with similar questions about the nature and scope of executive power in federations from perspectives that are more familiar in a manner that contributes to a deeper common understanding.

II. Federal design

At obvious risk of overgeneralisation, it is possible to identify two distinct approaches to the division of governing authority for federal purposes. They are described below as dualist (or thematic) and integrated (or functional). Each involves a very different treatment of executive power. Each has different implications for the lines of accountability for the exercise of executive power within and across jurisdictional lines. To complicate matters further, these two approaches are not necessarily mutually exclusive, even in design. And once the operation of federations in practice is taken into account, features become further intermixed in ways that may enhance the de facto exercise of executive power, without the corresponding provision for political or legal accountability likely to be associated with constitutional design.

As the term is used here, a dualist approach to federal design divides power vertically, in the sense that areas of government are assigned thematically to one or other of the spheres of government on a basis that assumes that each sphere will administer (or execute) its own legislation. Each sphere has a full set of institutions for the purpose: hence the label of « dualism ». The United States, Canada and Australia are examples. As these examples show federations in this category also may differ in a host of other ways, including the manner of the vertical division of power and the form of separation of powers for which the rest of the constitutional setting provides.

The division of legislative power, as the principal manifestation of sovereign authority, invariably is the most prominent under any federal model. In conditions of dualism, the legislative power allocated to each sphere is likely to include authority to raise revenues for its own purposes as well. By definition, however, consistently with this model, executive and, sometimes, judicial power also are divided between the respective spheres of government for federal purposes, implicitly if not explicitly. As a default position, these are likely generally to mirror the lines for the division of legislative power. In an extension of the logic of this model, there may be constitutional inhibitions on the extent to which one order of government can be required, or even invited, to exercise the executive power of another.

For the purposes of democratic accountability, each order of government in a dualist federation tends to be conceived as broadly self-sufficient, subject to constitutional provision to the contrary. Arrangements that cut across this proposition frequently include representation of the federated units in central institutions including, typically, a second chamber of a bicameral central legislature. Such a chamber may have considerable power within the central legislature but need not otherwise play a distinctively federal role. Co-operation across jurisdictional lines is common in such federations in practice. The occasion for co-operation may be stimulated by the federal division of powers but will be shaped by the dynamics of representative or, (where it exists), direct democracy, including the separation of powers. In most cases co-operation involves the executive branch, relies in significant degree on executive power and effectively extends executive authority. Some forms of co-operation detract from democratic accountability for the exercise of executive power and may raise concerns for that reason. A culture of co-operation may come less easily in dualist federations in any event, which are inherently more competitive in design.

In sum, therefore, in a dualist federation executive power is divided between the spheres of government along thematic lines. Whether and to what extent the question of the scope of executive power within each sphere is distinct from the scope of legislative power depends on the concept of executive power itself including, critically, the power of the executive to act without legislative authority. This issue is taken up in the next part. Political accountability for the exercise of executive power lies primarily within each order of government, in accordance with the arrangements for democratic government. Intergovernmental co-operation is likely to rely largely, if not exclusively on executive action and may cut across the principles and procedures for democratic accountability.

By contrast, in an integrated approach to federal design, power may be divided horizontally, along functional lines, empowering the sub-national sphere of government including, in some cases, local government, to administer or execute much central legislation as a matter of constitutional right. This approach to the division of power does not preclude a vertical division of powers as well. Every integrated federation also divides power on thematic lines; empowers the centre to administer some of its own legislation; and assumes that the sub-national sphere also will administer its own. The characteristic feature of a functional division of powers nevertheless ensures that administrative institutions are concentrated in the sub-national sphere. Where the power to tax also is divided along functional lines, taxation may be raised centrally but collected locally, necessitating constitutional provision for the distribution of the proceeds between spheres.

Germany serves as a prototype for federations of this kind. There are many other examples, however, including Austria, Switzerland, India and South Africa. In some cases, including Germany, the explanation for the adoption of a horizontal division of powers is historical reinforced, perhaps, by assumptions about the state-wide application of legislation. In other cases, the explanation for the choice variously lies in the challenges of accomplishing transition from a unitary to a federated state, in the expected demands of development, in a desire for a more centralised federation, or combinations of these. In some federations, of which India is an example, a horizontal division of powers is equated with the categorisation of legislative powers as concurrent. Federations that divide power along functional or horizontal lines typically are combined with parliamentary systems of government or, as in the case of Switzerland, a conciliar form of government. There may be a question about whether integration could work satisfactorily under a presidential system, with a more complete degree of separation between the legislature and the executive.

By definition, in an integrated approach to the federal division of power, the executive power exercisable by one sphere of government for the purposes of the federal division of power does not necessarily parallel its legislative power, even as a starting point for analysis. In lieu of, or in addition to, a question about the division of executive power for federal purposes is another, about what can be done by one order of government in executing the legislation of another. By definition again, under an integrated approach, mechanisms for accountability potentially cross jurisdictional boundaries. Whether and to what extent sub-national institutions are accountable to the centre or to their own voters for the execution of central laws depends on constitutional prescription and varies between federations. Federations of this kind may recognise a principle of federal loyalty, or good faith in dealings with each other. Some more recent federal type-systems, of which South Africa is an example, have equated this with a principle of co-operative government. In such federations, an (effectively) second chamber of the central legislature also may play an operative role. The German federation, again, is the principal case in point. The Federal Chamber, or Bundesrat, provides a vehicle through which the executive governments of the Lander play a role in considering and approving the central legislation that they will administer, in a form of intrastate co-operation.

It follows that, in all federations, there is a federal dimension to questions about the scope of executive power. The questions, and therefore the answers differ, however, both between and within the principal categories that I have outlined here. In federations designed as dualist, the scope of executive power depends largely although not necessarily exclusively on a division of power for federal purposes along thematic lines. In integrated federations, this division may bear less weight. In any event, in federations of this latter kind, it is relevant also to ask what can be done under the executive power that is constitutionally assigned to the sub-national sphere to implement central legislation. The distinction has implications for the understanding of mechanisms for accountability for the exercise of executive power.

Whatever the federal design, however, what can be done in the exercise of executive power in either sphere will also depend on the general conception of executive power. This is determined primarily by considerations of the separation of powers, informed by the rest of the constitutional context, of which federalism is only part. The general conception of executive power, including the extent to which it can be exercised without legislative authority, in turn will feed into the significance of the question of the division of executive power from a federal point of view.

III. Separation of powers

In any democratic state, federated or not, there is a constitutional conception of what can be done in the exercise of executive power. Often, executive power is conferred by legislation or, more generally, subsumed in the authority to implement legislation. Even in this familiar context, questions may arise about what the executive can lawfully do. The dispute in the United States over the extent of presidential discretion in enforcing immigration legislation is a case in point. In addition, however, there is an important sub-set of the conception of executive power that concerns the authority of the executive to act without legislative or specific constitutional authority. Executive power that is not derived from legislation but inheres in the executive qua executive presents an additional challenge for the division of power in federal-type systems and merits particular attention for this reason. The challenge is complicated further where inherent executive power comprehends a variety of matters with differing rationales. Thus, to take the example developed below, common law states may accept a conception of inherent executive power that combines powers in relation to the exercise of external sovereignty, of which treaty making and ratification and the commitment of troops to war are examples while, on the other with capacities of a kind ostensibly exercisable by ordinary persons, such as authority to contract and spend.

It may be that, in the 21st century, the core concept of executive power is broadly similar in all democratic states, precluding direct interference with rights and duties and authorising conduct of the day-to-day business of government. There are significant differences at the considerable margins, however, some of which also have a bearing on the dynamics of federalism. In each state, the conception of executive power is framed by a range of factors that might loosely be described as constitutional context. An important dimension of this context is the relationship between the legislative, executive and judicial branches of government and the powers that the Constitution authorises each to exercise, explicitly or implicitly. Another is the constitutional provision for fundamental rights.

At a deeper level still, constitutions tend simply to assume a conception of executive power, as a product of the historical experience of the state and its people, including that of the constitutional tradition of which they are part. In most common law states, for example, an underlying conception of executive power can be traced to the long evolution of the relationship between the Crown and Parliament, culminating in the revolutionary settlement of 1688, which retained the monarch as the embodiment of the state and limited but did not eradicate the prerogative. Outside the common law world, other formative influences have been at work: different historical experiences at different times and with different outcomes; different understandings of the sources of law including the role of legislation in legitimising the exercise of public power. Over time, at least in established constitutional systems, the implications of history for the conception of executive power are explored and developed by generations of theorists and jurists in ways that influence and are influenced by more formalised constitutional arrangements. The result, potentially, is a somewhat different understanding of the conception of executive power between states, masked by similarities in terminology.

The point may be illustrated by reference to the apparent homogeneity of the conception of executive power in common law states. Most obviously, there has been a degree of divergence between the British and United States constitutional traditions, stemming from the revolutionary break between the two at the end of the 18th century, differences in their constitutional arrangements in form and substance, and their different paths towards representative democracy, as parliamentary and presidential systems respectively. The initial point of divergence has been consolidated since by the intervening 200 years of the operation of the respective constitutional arrangements in practice, under different conditions, prompting different trajectories in judicial doctrines and different theoretical explanations. The claim could be substantiated by comparing many features of both traditions, but one aspect of institutional design serves the purpose. The respective dynamics of parliamentary and presidential systems inevitably shape the scope of executive power and the discourse associated with it. Relevant features for this purpose include the implications of direct, rather than indirect election of the executive in the United States; the capacity of the Parliament to override executive power in conditions of parliamentary supremacy; the effective autonomy of Congress in comparison with Westminster Parliaments; and the substitution of express presidential powers for powers in the nature of the prerogative, not least in empowering the President as commander in chief. Justice Jackson’s tripartite test for determining the scope of executive power in the Steel Seizures case has no obvious counterpart in analysis of executive power in the British constitutional tradition, where the assumptions that it makes about the relationship between the executive and the legislature do not apply. Albert Venn Dicey’s definition of the prerogative has no purchase in the United States where, indeed, Dicey is relatively unknown.

Even within the range of states that arguably derive their conception of executive power from the British tradition through processes of colonisation there now are differences in understanding of it. Most retain parliamentary systems, which they may continue to describe as « Westminster » in character. Explicitly or implicitly, these treat the settlement of 1688 as a definitive stage in the evolution of the scope of inherent executive power. Instinctively, all assume that the legislature can override general executive power. There has never been a clear and agreed analytical framework for the inherent executive power, however, nor for the rationale underpinning it. Different states may take paths of their own in resolving these questions for themselves, as problems arise from local practice. Invariably, also, there are particular differences in the constitutional experience of states in this tradition that have the potential to encourage divergent understanding of the conception of executive power. States with written Constitutions have a text on which interpretation can focus, unlike the United Kingdom from which the tradition derived. States that now are republics have an even greater incentive to abandon old analyses that drew on the royal prerogative or the capacities of the Crown as a legal person. In an interesting example of path-dependency, states in which power exercised by imperial authorities has been transferred to the executive under independence constitutions need to grapple with a broader conception of executive power than is accepted elsewhere. In those states that are also federations, an additional range of factors drawn from the competing interests of the centre and the constituent units may affect understanding of the conception of executive power as well.

This potential for divergent development of the concept of executive power in states in the British common law tradition is all the greater at a time when the concept of inherent executive power is under pressure in any event and, in that sense, is in a state of flux. Executive power in the nature of the prerogative, including the power to commit troops to armed conflict and to ratify treaties, is out of step with contemporary democratic expectations in relation to accountability, transparency and public consultation. Other categories of executive power, including the power to contract, whether characterised as within the prerogative or not, can be used to effect public policy goals binding future generations with limited parliamentary control. Whether through the evolution of convention, judicial interpretation or legislative change, different solutions to these problems are emerging from the experiences of these states, creating differences in the conception of executive power itself.

Features of a constitutional system that are associated with the separation of powers are the most obvious influence on the conception of executive power along this axis, but they are not the only ones. Others that are more subtle are easier to miss, however, for this reason. The impact of the choice between a generalist and a specialist court system for the purposes of organising adjudication serves as an example. While there are myriad variations, the latter typically mirrors a sharper conceptual distinction, not only between public and private law, but also between ordinary legal claims and claims about the constitutional validity of legislation. An apex court in a state in which generalist courts are the norm can deal with any legal issue properly raised before it, applying any applicable source of law including the Constitution, legislation and the common law (if any). In such a court, the constitutionality of executive action, even in the exercise of, say, a power to contract can be raised and resolved applying the full range of legal sources and tools. By contrast, a specialist constitutional court may be inhibited from dealing with a pure exercise of executive power at all which, in any event, may be characterised as an issue of administrative law or, (in the case of contract) as private law. Of course, arrangements for adjudication along specialist lines may merely reflect an established categorisation of legal questions. It may also serve to reinforce them, however. By contrast, a generalist court has a greater degree of flexibility in determining the boundaries of a private/public distinction between forms of executive power or even in eroding the distinction altogether.

The argument so far has sought to establish that the nature and scope of executive power in any federation involves examination of constitutional arrangements along two axes. One deals with the manner of the division of federal power. The other deals with the general conception of executive power primarily, although not exclusively, from the standpoint of the separation of powers. It has been convenient for analytical purposes to present the two dimensions of the problem separately here, but in reality they overlap. In any federation, the prevailing conception of executive power informs the organisation of power, including executive power, for federal purposes. Equally, the federal organisation of the state is part of the constitutional context that shapes the conception of executive power. Over time, this interdependence may increase, as the operation of the federation in practice raises an additional range of issues with which the conception of executive power must deal, causing it to be seen in a different light. While federalism interrelates with the conception of executive power in all federations, how this occurs and with what results varies in some degree. The next part illustrates how this relationship has shaped and continues to shape executive power in the federation of Australia.

IV. Australia

A. Federal design

The Australian federation was formed in 1901 as an agreed means of bringing together six established self-governing British colonies in a single polity. It thus had many, although not all, the characteristics of federation by aggregation. In an important exception, colonial status within the British Empire at the time of federation meant that the colonies that became States never possessed full sovereignty and that the Commonwealth itself lacked powers associated with external sovereignty for decades after federation took place.The continued, unbroken relationship with the United Kingdom also required incorporation of the monarchy in the new Constitution. These factors have had implications for the executive power of both the Commonwealth and the States in the Australian federation, not the least of which has been the expansion of the scope of executive power in section 61 as imperial authority withdrew. Otherwise, however, as in most cases of aggregation, the States retained their existing Constitutions, institutions and laws, subject to provisions to the contrary in the Commonwealth Constitution. Apart from the inroads necessarily made by federation into the scope of State power, these were not onerous. The Constitution established only federal institutions, leaving those of the States largely untouched. The Constitution also left rights protection to the institutions of the several spheres of government, in the exercise of their respective powers.

The expectation that each order of government executes its own legislation is evident on the face of the Constitution. Subject to the role of the High Court of Australia as the final court of appeal in both federal and state jurisdiction, each order of government also provides for adjudication on matters arising in its own sphere of authority. The Constitution divides legislative, executive and judicial power between the spheres of government to these ends, by specifying the powers of the Commonwealth, leaving the residue to the States. The division of legislative power is spelt out in some detail, in more than 40 heads of power that are made available to the Commonwealth Parliament. Most, although not all, of these are held concurrently with the States, subject to the paramountcy of Commonwealth law. They include an incidental legislative power in section 51(XXXIX) that enables supporting legislation with respect to « matters incidental to the execution of any power vested by this Constitution […] in the Government of the Commonwealth ». The heads of federal jurisdiction also are relatively detailed.

The treatment of executive power, on the other hand, is much less specific. Section 61 provides:

The executive power of the Commonwealth is vested in the Queen and is exercisable by the Governor-General as the Queen’s representative, and extends to the execution and maintenance of this Constitution, and of the laws of the Commonwealth.

Some implications of the section nevertheless are clear. The executive power of the Commonwealth is distinct from that of the States. In terms of content, it clearly involves the execution and maintenance of Commonwealth law. What else it involves depends on the significance of the word « extends », the ambit of authority to execute and maintain the Constitution, the consequences, if any, of the investiture of power in the Queen and the very conception of « executive » power.

The federal division of fiscal authority also is part of the story of the federal division of executive power in Australia. The Constitution allocates extensive power to impose taxation to each of the spheres of government, on the dualist assumption that each government would tax for its own purposes. In an important exception, however, the power to impose duties of customs and excise was conferred exclusively on the Commonwealth, depriving the new States at the moment of federation of one of their most significant sources of revenue. As a result, from the outset, there was a degree of fiscal imbalance, for which the Constitution attempted to make remedial provision. One section, which became a dead-letter in the early years of federation, required the regular distribution of Commonwealth « surplus » revenue to the States. Another provided simply for the payment of financial assistance by the Commonwealth to any State on such terms and conditions « as the Parliament thinks fit ». Over time, the fiscal imbalance worsened, as the Commonwealth acquired a de facto monopoly over income tax and the High Court interpreted the exclusive Commonwealth power to impose duties of excise as encompassing, in effect, all taxes on goods. From the mid-20th century at the latest the Commonwealth had very considerable sources of revenue at its disposal, while the States were dependent on transfers, pursuant to section 96. This mismatch of legislative power and resources encouraged Commonwealth reliance on executive power to unilaterally extend its authority into areas of State responsibility, creating a recurring flashpoint for litigation that has been significant in shaping the scope of the power.

One final aspect of Australian federal design that has affected both judicial doctrine and political practice in relation to executive power is the Senate. The Senate is the second chamber of the bicameral Commonwealth Parliament. It represents all the original States equally, is directly elected and has almost co-equal authority with the House of Representatives. It is rare for a government to have a majority in the Senate, providing an additional incentive for governments to avoid legislation in favour of reliance on executive power where they can. On the other hand, avoidance of the legislature raises concerns for federalism as well as separation of powers when the Senate is the principal formal institution of shared rule.

B. Separation of powers

The Australian Constitution does not define the concept of executive power that is divided between the Commonwealth and the States by section 61. While other constitutional provisions detail aspects of executive power for particular purposes, they provide almost no assistance with the meaning of the generic term. Like the companion concepts of legislative and judicial power in sections 1 and 71 of the Constitution, the meaning of executive power is simply assumed, leaving uncertainties to be resolved through judicial interpretation.

Given the provenance of the Australian Constitution, one source of insight into the meaning of the concept of executive power is British constitutional law and practice. The potential relevance of British experience is heightened by Australian adaptation of key British institutions including constitutional monarchy and parliamentary responsible government broadly, but by no means entirely, along Westminster lines. The British conception of executive power might add to understanding of section 61 in at least three ways. First, executive power in the United Kingdom extends well beyond the execution of legislation to include prerogative power. Secondly, the common law also recognises the inherent authority of the executive to carry out a range of other actions, broadly of a kind that might be performed by any legal person, as long as they are not contrary to law. This anthropomorphism assumes that the executive can be equated with a legal person for this purpose, sometimes by relying on the personhood of the Crown. If either of these possibilities applies in Australia, the immediate result is to extend the concept of executive power to which section 61 refers beyond execution of the Constitution and statute law. Thirdly, in a feature of a different kind, executive power in the British parliamentary tradition can always be overridden by Acts of the sovereign legislature so long, at least, as the intention to do so is sufficiently clear.

The relevance of the British conception of executive power in Australia is necessarily tempered by constitutional differences between the two states, however. Most obviously, in Australia the concept of executive power is embedded in an entrenched constitution enforced through judicial review. The Commonwealth Parliament is not sovereign in the sense generally accepted for the Parliament at Westminster and could not override specific powers conferred on the executive by the Constitution, whatever the relationship between legislation and section 61. Equally significantly, the Constitution in which executive power is embedded protects approaches to the organisation of public power that in some respects are quite different to those in the United Kingdom, not least of which is the federal system itself. Thus, for example, if the concept of executive power under section 61 includes significant inherent executive power, divorced from statute, it becomes necessary to determine the basis on which power of this kind is divided between the Commonwealth and the States.

An alternative source of insight into the concept of executive power is the Constitution of the United States, on which the Australian Constitution was loosely modelled. The profound difference, for this purpose, between a presidential and a parliamentary system means that the United States offers no guide to the precise conception of executive power. Nevertheless, the similarities in organisation of the first three parts of each Constitution, dedicated to the legislature, executive and judicature respectively, suggests that the questions that arise in Australia might also be analysed through a separation of powers lens. This possibility has been realised to the extent that the Australian Constitution has been held to provide for a three-way separation of powers, albeit one that makes allowance for the exigencies of parliamentary government. It follows that there are enforceable limits on what can be done by the Commonwealth in the exercise of executive power, the demarcation of which falls to be determined by reference to the text and context of the Australian Constitution. The concept of State executive power requires a somewhat different analysis, involving the State as well as the Australian Constitutions.

C. Breadth and depth

The scope of federal executive power under section 61 has been a familiar legal battleground since federation. Over time it became accepted that, the final words of the section notwithstanding, executive power was not necessarily dependent on legislative or express constitutional authority: that, in other words, the Commonwealth executive enjoyed some inherent executive power. The extent of inherent power has remained contested, however, from the standpoint of both federalism and separation of powers. Nor has it ever been entirely clear how this development relates to section 61. On one, now older view, prerogative and/or general executive power derived from the common law and were merely subsumed by the section. On another, the function of ideas about executive power derived from the common law constitutional tradition is to assist to inform the meaning of section 61 in constitutional context. On yet another, the references in section 61 to the Queen or to the « execution and maintenance » of the Constitution might be interpreted to authorise much of the action that could be taken in the exercise of inherent executive power elsewhere.

Australian discourse has sometimes described the problem of determining the scope of the executive power of the Commonwealth in terms of breadth and depth, broadly reflecting the two axes of executive power in federations with which this chapter began. Depth thus refers to the scope of executive vis-à-vis legislative power and breadth refers to the ambit of the executive power of the Commonwealth vis-à-vis that of the States. Most litigation has focussed on one or the other, encouraging discrete treatment of the two dimensions of the problem. Thus in Barton v. Commonwealth, in which the plaintiff challenged an Australian request to extradite him from Brazil, the principal issue was the relationship between prerogative power and statute and the federalism dimension was, properly, assumed to be satisfied. Conversely, in the Australian Assistance Plan (AAP) case, argument focussed on whether expenditure on Regional Councils for Social Development fell within the powers of the Commonwealth, without seriously questioning whether the executive could spend public funds to this end without substantive supporting legislation.

The AAP case was complicated for present purposes because the Court was divided over whether the power of the Commonwealth to engage in spending programs of this kind was unlimited by federal considerations and whether, in any event, it derived from the requirement for appropriation in section 81 or relied, at least in part, on the executive power. The reasons of Mason J offered an influential middle ground. On this view, the executive power would support engagement in spending programs in areas of Commonwealth constitutional responsibilities, which could be ascertained from the distribution of legislative powers and from the « existence and character of the Commonwealth as a national government ». The latter gave the Commonwealth « a capacity to engage in enterprises and activities peculiarly adapted to the government of a nation and which cannot otherwise be carried on for the benefit of the nation ». Notably, the judge who took this view held that the Australian Assistance Plan was invalid.

Questions about the nature and scope of executive power in Australia have been raised in other contexts as well, further complicating an already complex jurisprudence. The executive power to contract without adequate parliamentary authority has been challenged at the State level as well, with results that have a bearing on problems arising in the Commonwealth sphere, despite the more « unitary » context. Litigation challenging the outcomes of intergovernmental schemes has established that the Commonwealth executive has power to enter into agreements with the States, subject to constitutional limits that remain to be explored. It should be noted in passing that this is an aspect of executive power peculiar to a federal setting. Long-running lines of authority about whether and when the respective spheres of government could enact legislation that applied to the executive branch of the other led to decisions that tended to assume inherent executive power to contract in the course of, for example, drawing a distinction between State legislation that affected the « capacities and functions » of the Commonwealth, and legislation regulating the transactions in which the Commonwealth executive « chose to engage ». The former was precluded but the latter was not, subject to Commonwealth legislation to the contrary.

D. A compound conception of federal executive power

As long as the breadth and depth of executive power were treated as discrete questions in Australian constitutional litigation, the latter drew primarily on modes of analysis derived from the British constitutional tradition, with outcomes that were broadly equivalent. From the turn of the 21st century, however, a wave of cases raising problems of executive power began what with hindsight can be seen as a new turning. Executive power in the Australian federation now is a compound conception in which considerations of federalism and separation of powers influence each other in ways that are determined by reference to the Australian Constitution as a whole.

Oddly, the starting point was a decision of the Full Court of the Federal Court of Australia in which federalism was not in issue at all. Instead, the question for the Court in Ruddock v. Vadarlis was the validity of actions of the Commonwealth executive in intercepting and expelling non-citizens arriving by boat in reliance on inherent executive power. British authority on the scope of the prerogative for this purpose was equivocal, as the dissenting reasons of Black CJ showed. For the majority judges, however, this was not determinative. The executive power of the Commonwealth was not a « species of the royal prerogative » but a « power conferred as part of a negotiated federal compact expressed in a written Constitution distributing power between the three arms of government […] and […] as to legislative powers, between the polities that comprise the federation ». Construed in context, section 61 authorised the action here, which was « central to […] sovereignty » and thus attributable to « maintenance of the Constitution ». The reasoning by which this result was reached drew extensively on the notion of nationhood previously used primarily to delimit the federal dimensions of the power.

Federalism was directly in issue in the three cases that followed, later in the decade. The first of these, Pape v. Federal Commissioner of Taxation, narrowly upheld the validity of economic stimulus legislation in the wake of the global financial crisis, on the grounds that a short-term emergency of this kind attracted the nationhood component of the executive power, which could then be supported by an exercise of the incidental legislative power. Importantly for present purposes Pape also confirmed, however, that any executive power to spend was limited by considerations of federalism; that its scope depended on the interpretation of section 61, in the context of the Constitution as a whole; and that the requirement for appropriation played no necessary role in the resolution of the problem, prescribing instead the technical procedure for the release of moneys from the Consolidated Revenue Fund. The reasoning in Pape made it likely that further challenges would follow as they did, in the two Williams cases.

The issue in both cases involved spending pursuant to a contract. In pursuance of a National Schools Chaplaincy Program, the Commonwealth had entered into a Funding Agreement with the Scripture Union Queensland (SUQ) to provide funding for chaplaincy services in Queensland schools that expressed interest in a chaplaincy placement. Individual contracts were concluded between the Commonwealth and SUQ in relation to the placement of each Chaplain. The plaintiff was the father of children at one of the schools in question, who objected to having a chaplain at the school. The Program was undertaken solely in exercise of the executive power, subject to a general appropriation in the budget legislation appropriating moneys for the « ordinary annual services of government » that specified the « outcome » as « Individuals achieve high quality foundation skills and learning outcomes from schools and other providers ». The regulatory framework for the program was provided by executive Guidelines, which were subject to frequently change.

In the first Williams case, the High Court held by a majority of 6-1 that the Funding Agreement was not a valid exercise of the executive power of the Commonwealth and was not supported by section 61. In itself this was not a surprising outcome, although it was hard fought by the defendant Commonwealth. The case is significant for the reasoning of four of the majority Justices, who reached their decision on the basis that the contract and associated spending were unconstitutional because they were unsupported by legislation, rather than because they exceeded Commonwealth, as opposed to State executive power. In other words, the majority held that the contract did not fall within the constitutional conception of executive power at all. The other two majority Justices avoided decision on this issue, holding instead that the contract fell outside the scope of Commonwealth executive power, measured by reference to the somewhat more orthodox considerations of the contours of legislative power augmented by the demands of nationhood.

There was no single majority opinion; a familiar situation in Australian practice. As a result, there are some significant differences between the majority reasoning. There was enough in common between the three sets of reasons, however, to draw the following conclusions from them for the nature and scope of Commonwealth executive power. The text of section 61, interpreted in constitutional, including historical context, is the primary point of reference. The scope of Commonwealth executive power is not determined by the scope of the prerogative or common law executive power in the United Kingdom, although this may inform the understanding of section 61. Insofar as reasoning about the scope of inherent executive power in other common law states rests on an analogy with the capacities of a legal person, reinforced by the location of executive power in the Crown, it has been rejected for Australia. The Constitution creates the Commonwealth (and, by inference each State) as a polity and distributes functions between its branches in a manner that, ultimately falls for determination by the Court. Construed in context, Commonwealth executive power necessarily is a composite concept, in which understanding of what the executive branch can do without legislative authority is shaped by considerations of both separation of powers and federalism, in the light of constitutional text and structure. Federalism indicators that influenced the reasoning of the Justices in Williams included the potential for a broad executive power, underpinned by the incidental legislative power, to undermine the constitutionally entrenched division of legislative powers, especially on a matter that clearly lay within State authority and could not conceivably be a candidate for the « nationhood » power; the presence in the Constitution of section 96, authorising Commonwealth spending in areas beyond its legislative powers through grants to the States on such conditions as the Parliament « thinks fit »; and the potential for a program based solely on the executive power to avoid the need for Senate consideration and approval.

The sequel to the first Williams decision showed how the now compound conception of federal executive power might have benefits for representative democracy as well as federalism. The immediate response to Williams was legislation to provide a loose subordinate statutory base for the National Schools Chaplaincy Program and more than 400 other executive spending programs deemed to be at risk. Despite the somewhat provocative nature of this solution, which remains to be tested, any legislative underpinning for the scheme was vulnerable to challenge by reference to the federal division of legislative power. In Williams (No 2), a challenge to the validity of the legislation was upheld by a unanimous High Court on the ground that there was no head of legislative power to support it. Insufficiently daunted, the government now gave the School Chaplains Program effect through conditional grants to the States pursuant to section 96. While this was a hollow victory for the plaintiff, it was at least consistent with the federal framework. The intergovernmental agreement setting out the terms of the grant also is more stable in policy terms, if only because it involves two parties, than the departmental guidelines had proved to be. Further, the new subordinate legislative basis for federal spending programs exposes them to scrutiny by the Senate Standing Committee on Regulations and Ordinances, which now insists that Ministers identify the federal head of power on which they rely, when new spending regulations are made. And in yet another twist, the attention aroused by the transformation of the program into a conditional grant to the States caused the Senate Standing Committee for Scrutiny of Bills to begin to take a belated interest in the way these arrangements work.

E. Unfinished business

It seems likely that understanding of the nature and scope of executive power in Australia will continue to build on the doctrinal framework laid down in the first Williams case. The Court rejected the Commonwealth’s attempt to reopen these issues in Williams (No 2). Political practice is beginning to adjust to a more limited ambit for inherent executive power. It may be that the doctrine is not entirely secure, however. The terms of the joint reasons in Williams (No 2) were sufficiently carefully phrased to preserve both the strands of reasoning over which the six-member majority divided in the earlier case. And in a post-Williams decision, on a matter entirely within the federal sphere of responsibility, one more recent member of the Court went to some lengths to confine the authority for which the cases stand.

The doctrine is vulnerable also, however, because aspects of it remain unclear. The existence of Commonwealth executive power in association with a valid Commonwealth statute or pursuant to particular provisions of the Constitution is straightforward enough. In these instances, also, questions about federalism are automatically resolved. Commonwealth claims to inherent executive power present a more complex case. Williams suggests several bases on which they might be resolved. In some cases the executive might exercise powers « in the nature of the prerogative » that are properly attributable to the Commonwealth; in others executive action might be justified by « the character and status of the Commonwealth as the national government »; in others again, the executive has power to act in the course of administering the Departments of State. The federal character of the Constitution plays a role of some kind in each of these categories. Nevertheless, their scope and the boundaries between them are by no means clear. In particular, it remains to be determined which contracts and spending programs require supporting legislation and which do not. This is an issue on which comparative experience might well assist.

In addition, it remains to be determined whether the same limits on the capacity of the Commonwealth executive to act without legislation affect the executive power of the States. At first glance, the dependence of the Williams reasoning on the context of the Australian Constitution, including its federal features, makes extension to the States unlikely. Separation of powers also is a much less prominent feature of the Constitutions of the States. On the other hand, the Australian Constitution recognises both the Commonwealth and the States as polities, providing a basis on which anthropomorphic reasoning can dismissed for both. And there are precedents in Australia for the extension to the States of the effect of doctrines governing the position of the Commonwealth under the Australian Constitution. This can be expected to happen, in one way or another, in connection with executive power as well.

V. Conclusions

This study of executive power in federal systems suggests three points of particular interest for comparative federalism.

First, the arrangements by which power is organised and limited in a working federation are interdependent with the rest of the constitutional system, which also may evolve over time to reflect the modalities of federalism. Thus, while legislative and executive powers inevitably are divided between the orders of government in any federation, how they are divided and the manner of their exercise depends on the constitutional and legal setting in which they are embedded. Whether executive power presents itself as a distinct issue may depend on the extent to which there is a conception of executive power independent of legislation or of the scope of legislative power. Where, on the other hand, the constitutional setting acknowledges substantial inherent executive power it presents an additional set of problems for federalism to resolve. The Australian case study shows how, in these conditions, the conception of executive power may adapt to the federal setting.

Reflection on these dynamics suggests the second point. Key features of the constitutional setting in which federalism is embedded generally have their genesis in a unitary system of government. This may particularly be so, for example, in relation to the structure and power of the three branches of government, the legislature, executive and judiciary, arrangements for shared rule aside. The relationship between these branches typically is underpinned by historical experience, whether local or derived, given continuing relevance through constitutional theories. The political culture in which they have evolved is not necessarily conducive to the power-sharing and consensus-building that federalism requires. In these circumstances, there is potential for a clash between the two sets of principles on which a federal constitution rests, each of which may be resistant to change. The Australian case study illustrates these dynamics in relation to the concept of executive power in one federation. It also shows how, in those particular conditions, a modus vivendi of sorts has been reached.

Third, as a logical corollary, the nature and scope of executive power differs in some respects between all federations, depending both on federal design and the constitutional setting more generally. One aim of this essay has been to suggest an approach to understanding these differences and evaluating their significance for comparative purposes. The exercise is complex, because some differences run deep. It nevertheless is worthwhile. The conditions in which government operates in the early 21st century tend to augment executive power, externally and internally. This phenomenon affects all states but has particular implications for states that also divide governing authority for federal purposes. There are insights to be gained into how such pressures might be both accommodated and resolved from a comparison of federal experiences that is adequately informed.

Cheryl Saunders

Cheryl Saunders is a Laureate Professor Emeritus at Melbourne Law School, Australia, President Emeritus of the International Association of Constitutional Law and a former President of the International Association of Centres for Federal Studies. She has written widely on Australian and comparative constitutional law, including federalism and intergovernmental relations.