The Non-Abdication Rule in Canadian Constitutional Law Mark Mancini I

The Non-Abdication Rule in Canadian Constitutional Law

Mark Mancini*

I. INTRODUCTION

Delegation of legislative power from Parliament to other actors in the federal government, what scholars refer to as lateral delegation, runs rampant in Canada.1 Independent agencies2 wield

2020 CanLIIDocs 693 * BA, JD, LLM; National Director, Runnymede Society.

1 See generally Lorne Neudorf, “Reassessing the Constitutional Foundation of Delegated

Legislation in Canada” (2018) 41:2 Dal LJ 519. I critically reference Neudorf’s piece

throughout the paper as I am offering a distinct account of delegation limits. See also

John Mark Keyes, Executive Legislation, 2nd ed (Toronto: LexisNexis, 2010) at 103–

147. Keyes’ book in particular is a magnum opus of delegated legislation, speaking to

various facets of the problem of delegated legislation in Canada, including its technical

aspects.

2 I use this term loosely because there is no widely adopted definition of an independent

agency. More accurately, there is a spectrum of independence in which some agencies

have more independence from the executive branch and legislatures than others.

Generally, at the federal level, I am speaking of agencies like the Canadian

Transportation Agency, the Canadian Human Rights Commission, the Canadian Human

Rights Tribunal, the Canadian Energy Regulator, the Immigration and Refugee Board of

Canada, and the Canadian Radio-television and Telecommunications Commission. See

Matthew Flinders, Delegated Governance and the British State: Walking without Order

legislative power3 and are insulated to a certain degree from political influence. While functional concerns often motivate delegation of this sort,4 that does not mean that it is formally consistent with the organizing principles of the British Westminster parliamentary tradition, of which

Canada is a part. This question of consistency is not of purely academic or hypothetical interest.

Important regulatory legislation is made primarily at the Cabinet and agency level, with five

2020 CanLIIDocs 693 (Oxford: Oxford University Press, 2008) at 5 for the idea of independent agencies on a

spectrum of independence; Lorne Sossin, “The Puzzle of Independence for

Administrative Bodies” (2008) 26:1 NJCL 1 for general discussion and definitional

terminological considerations.

3 In defining legislative power for the purposes of this paper, I adopt the Supreme Court of

Canada’s approach to defining legislation under its “prescribed by law” jurisprudence.

See Greater Vancouver Transportation Authority v Canadian Federation of Students—

British Columbia Component, 2009 SCC 31 at para 64, [2009] 2 SCR 295 [GVTA]:

So long as the enabling legislation allows the entity to adopt binding

rules, and so long as the rules establish rights and obligations of general

rather than specific application and are sufficiently accessible and

precise, they will qualify as ‘law’ which prescribes a limit on a Charter

right.

4 These reasons include a desire for efficiency, expertise, and independence. See generally

Frans F Slatter, Parliament and Administrative Agencies (Ottawa: The Law Reform

Commission of Canada, 1982) at 8–19.

times more delegated legislation than primary legislation.5 At the federal level, there are 201 departments and agencies listed by the Government of Canada,6 many of which have the power to issue delegated legislation. However, volume is not the only measure of the significance of delegated legislation.7 Important issues of political significance affecting individual rights and interests are also decided by administrative agencies. Indeed, the Supreme Court of Canada has remarked that regulations are the “life blood” of the administrative state.8

One could say there is no legal impediment, such as a doctrine of non-delegation, to stop 2020 CanLIIDocs 693 this mass transfer of legislative power away from Parliament.9 Parliament is sovereign10 and can

5 Neudorf, supra note 1 at 521.

6 Government of Canada, “Departments and Agencies” (last modified 24 January 2020),

online: , archived: .

7 See generally Neudorf, supra note 1.

8 Alberta v Hutterian Brethren of Wilson Colony, 2009 SCC 37 at para 40, [2009] 2 SCR

37.

9 But see (Alyn) James Johnson, “The Case for a Canadian Nondelegation Doctrine”

(2019) 52:3 UBC L Rev 817; Neudorf, supra note 1.

10 Guy Régimbald & Dwight Newman, The Law of the Canadian Constitution (Markham:

LexisNexis, 2013) at 70: “Due to the constitutional limits imposed on the principle, it

would be in error to confuse the principle of Parliamentary sovereignty with that of

Parliamentary supremacy…[I]n Canada, the division of powers and the Charter make it

preferable to use the term ‘Parliamentary sovereignty.’”

thus “make or unmake any law whatever.”11 Subject to any regent constitutional limitations, the upshot for delegation of law-making power is clear, at least in theory. There is no general, ex ante restriction on the scope of law-making powers that Parliament can laterally delegate to the executive or other law-making bodies.12

Yet, oddly, the Supreme Court of Canada has adverted to other limits on delegated power. Notably, it has said that Parliament cannot “abdicate” its law-making powers through delegation by giving powers to those ineligible to receive them.13 Canadian case law seems to 2020 CanLIIDocs 693 equate abdication with the withdrawal of “control” over the recipient of delegated power, which

11 AV Dicey, Introduction to the Study of the Law of the Constitution, 3rd ed (London:

MacMillan & Co, 1889) at 38.

12 Bogdan Iancu, Legislative Delegation: The Erosion of Normative Limits in Modern

Constitutionalism (Heidelberg: Springer, 2012) at 1: “[T]he English constitution…resists

normative limitations on the legislature.”

13 See In Re George Edwin Gray, (1918) 57 SCR 150, 1918 CanLII 533 [Gray cited to

SCR]; Attorney General of Nova Scotia v Attorney General of Canada, [1951] SCR 31 at

44, 1950 CanLII 26 [NS Interdelegation]. As noted in NS Interdelegation, a provincial

legislature cannot “create and endow with its own capacity a new legislative power not

created by the Act to which it owes its existence” (ibid at 38, quoting In the matter of The

Initiative and Referendum Act being Chapter 59 of the Acts of Legislative Assembly of

Manitoba 6 George V. v Manitoba, [1919] UKPC 60, 48 DLR 18 [Re Initiative and

Referendum Act cited to DLR]).

is usually the executive.14 Coupled with this is an implicit recognition that the mere possibility of the revocation of the delegated power is not enough to render a delegation constitutional.15 While some scholars have addressed the issue of delegation, none have addressed what “abdication,” in particular, might mean in its legal and institutional forms.16 Some have even discarded the

14 See the Reference as to the Validity of Regulations in relation to Chemicals enacted by

Order in Council and of an Order of the Controller of Chemicals made pursuant thereto, 2020 CanLIIDocs 693 [1943] SCR 1 at 18, 943 CanLII 1 [Chemicals Reference]; Gray, supra note 13 at 176;

Powell v Apollo Candle Co. Ltd. (New South Wales), [1885] UKPC 5, 10 App Cas 282 at

291. In the Westminster tradition, the “executive” largely refers to the Cabinet and

individual ministers. These people are also members of Parliament, drawing a clear

distinction between the American separation of powers where the executive does not sit

in the legislative chambers. In turn, the executive is responsible to Parliament, and

Parliament controls the executive.

15 NS Interdelegation, supra note 13 at 50: “The power of revocation might in fact become

no more feasible, practically, than amendment of the Act of 1867 of its own volition by

the British Parliament.”

16 The leading article to explore this issue is John Willis, “Administrative Law and the

British North America Act” (1939) 53:2 Harv L Rev 251. For rare exceptions of scholars

wrestling with issues of delegation particularly as they pertain to regulatory policy, see

Harry W Arthurs, “Regulation-Making: The Creative Opportunities of the Inevitable”

(1970) 8:3 Alta L Rev 315; Alice Woolley, “Legitimating Public Policy” (2008) 58:2

UTLJ 153; Roderick A Macdonald, “Understanding Regulation by Regulations” in Ivan

normative importance of non-abdication as a restriction on delegation in certain circumstances.17

This is particularly so with the rise of independent agencies that are insulated to varying degrees from parliamentary control—“structural heretics” in the constitutional structure.18

Accordingly, this paper analyzes what it could mean for Parliament to “abdicate” its legislative power. Specifically, I offer an argument distinct from the existing literature: the

Canadian non-abdication rule puts restrictions on who may receive delegated power, rather than on the scope of delegated power, as in the American non-delegation doctrine. This means that 2020 CanLIIDocs 693 Parliament can delegate power as it sees fit. The relevance lies in who receives this broad power.

When Parliament delegates to an independent agency, it must exercise control over the delegated law-making by that agency in order to ensure that it does not abdicate its legislative power. I locate this ex post control in the idea, fundamental to the Westminster tradition,19 that Parliament

Bernier & Andrée Lajoie, eds, Regulations, Crown Corporations and Administrative

Tribunals (Toronto: University of Toronto Press, 1985) 81 at 119.

17 None other than Bora Laskin called the non-abdication constraint “at most something to

talk about” in war time although he conceded that “in peace time it may be something to

act upon” (Bora Laskin, “Administrative Law—War—Subdelegation of Power without

Express Statutory Authorization” (1943) 21 Can Bar Rev 141 at 144); see also Keyes,

supra note 1 at 113.

18 See generally JE Hodgetts, Pioneer Public Service: An Administrative History of the

United Canadas, 1841–1867 (Toronto: University of Toronto Press, 1956) at 143–47.

19 The Westminster tradition of Parliament refers to “a Parliament with effectively unified

executive and legislative branches and the institution of ministerial responsibility”

exercises control over the executive and its delegates with a concomitant prohibition on the creation of new law-making bodies unaccountable to Parliament. The result is a judicially enforceable doctrine that limits who may receive delegated power.

In Part II, I contrast what the Supreme Court of Canada has said about delegation with the non-delegation doctrine articulated by the Supreme Court of the United States. I do this to demonstrate that Canada’s restrictions on recipients of delegation are different from the

American restrictions on the scope of delegation, owing to the separate constitutional 2020 CanLIIDocs 693 arrangements in each country.

Then, in Part III, I argue that the non-abdication rule in Canada attaches to the recipient of delegated power rather than to the scope of delegated power, as in the United States. First, I argue that restrictions on the scope of delegation are ill-suited to the Canadian system of government, which is based on parliamentary sovereignty and permits Parliament to pass any law whatsoever (subject to constitutional restrictions). Second, since there are no relevant constitutional restrictions on parliamentary sovereignty when it comes to delegated power, the scope of power cannot be limited. Third, given the prominence of parliamentary sovereignty, the non-abdication doctrine applies to restrict the delegation of legislative power to independent agencies, such that Parliament cannot create a new law-making body without controlling it.

Finally, in Part IV, I apply these observations to two case studies raising the non- abdication argument: the recent Reference re Greenhouse Gas Pollution Pricing Act20 from the

(Benedict Sheehy & Don Feaver, “Re-Thinking Executive Control of and Accountability

for the Agency” (2016) 54:1 Osgoode Hall LJ 175 at 185, n 30).

20 2019 SKCA 40, 9 WWR 377 [Saskatchewan Reference].

Court of Appeal for Saskatchewan and the Reference re Pan-Canadian Securities Regulation21 from the Supreme Court of Canada. Specifically, I argue that the non-abdication rule has no force in the former case because the argument was raised with respect to the scope of the delegation. However, it does have force in the latter case because the delegation was to a new law-making body, contrary to the restriction on recipients prescribed by the non-abdication rule.

I conclude the section by offering some complications and implications associated with the argument from an institutional perspective. 2020 CanLIIDocs 693

II. DELEGATION RULES: CANADA AND THE UNITED STATES

A. CANADIAN DELEGATION RULES

The Supreme Court of Canada has said little about what it might mean for Parliament to abdicate its legislative power, assuming that such an abdication would ever be possible.22 Nonetheless, it is important to begin by describing what the Supreme Court of Canada has said about abdication, sparse though the examples may be. This description will introduce the discussion in the next part about the different delegation limit that applies in the American context of separated powers, where the scope of the delegation is the major concern.

Delegation cases were litigated early in Canada’s history. An important early case is

Hodge v. The Queen,23 a decision of the Judicial Committee of the Privy Council. Hodge

21 2018 SCC 48, [2018] 3 SCR 189 [Pan-Canadian Securities Reference].

22 Willis, supra note 16 at 254–55, noting that determining what non-abdication means is

“generally regarded as insoluble.”

23 [1883] UKPC 59, (1883) 9 App Cas 117 [Hodge].

involved a constitutional challenge to the authority of the provincial legislature of Ontario to delegate the power to issue tavern licences to the License Commissioners of Toronto. This case may not appear to deal with legislative power as the issuance of tavern licences appears to be more of an executive or judicial power.24 Nonetheless, the assertion was that provinces were mere delegates of power and that the province of Ontario could not “sub-delegate” power to the

License Commissioners of Toronto. The Judicial Committee of the Privy Council rejected this assertion. Noting that “[i]t was argued at the bar that a Legislature committing important 2020 CanLIIDocs 693 regulations to agents or delegates effaces itself,” the Judicial Committee of the Privy Council ultimately concluded that a legislature in Canada could delegate a wide scope of power, and later revoke this delegation if it became problematic.25 Bolstering this conclusion was the fact that, for the Judicial Committee of the Privy Council, “[i]t [was] obvious that [the conveyed] authority

[was] ancillary to legislation.”26 At the same time, the Judicial Committee of the Privy Council also noted that delegations are “matters for each Legislature, and not for courts of law, to decide.”27 Here is the nascent idea that courts cannot attach limits to the scope of power delegated by a competent legislature.

Further limits on delegation followed. In the matter of The Initiative and Referendum Act being Chapter 59 of the Acts of Legislative Assembly of Manitoba 6 George V. v. Manitoba, the

Judicial Committee of the Privy Council seemed to adopt a more nuanced understanding of the

24 Willis, however, deals with the issue as one of delegated legislative power. See Willis,

supra note 16 at 252 under the heading “Delegation of Legislative Power.”

25 Hodge, supra note 23 at 12.

26 Ibid.

27 Ibid.

limits of delegation. The Judicial Committee of the Privy Council rejected an attempt by the

Manitoba Legislature to permit the adoption of citizens’ initiatives by a majority vote without passage through the legislature and without Royal Assent. The problem with the referendums was an early example of what the Judicial Committee of the Privy Council would later call abdication, though not stated in these terms at that time. According to Viscount Haldane, a legislature may not “create and endow with its own capacity a new legislative power not created by the Act to which it owes its own existence.”28 This was recognized despite the Judicial 2020 CanLIIDocs 693 Committee of the Privy Council’s acknowledgement that Parliament could, indeed, delegate its powers.29 However, what Re Initiative and Referendum Act stands for is the establishment of some limit to that power. The limit appears to be a prohibition against the creation of a law- making body that would rival Parliament; one that Parliament cannot, in some way, control. This limit was not mentioned in Hodge but was implied in another Judicial Committee of the Privy

Council case, The Queen v. Burah,30 in relation to the Indian Constitution.

Another Judicial Committee of the Privy Council case, Shannon v. Lower Mainland

Dairy Product Board,31 concerned the constitutional validity of a legislative scheme that delegated power from the British Columbia Legislature to the Cabinet of the province of British

Columbia. Under this power, the Cabinet created the Dairy Product Board. The appellants challenged the constitutionality of the Board on a number of grounds, including that “it [was] not

28 Re Initiative and Referendum Act, supra note 13 at 25.

29 Ibid.

30 [1878] UKPC 1, 3 APP CAS 889.

31 [1938] UKPC 54, [1938] 4 DLR 81 [Shannon cited to DLR].

within the powers of the Provincial Legislature to delegate so-called legislative powers to the

[Cabinet]”32 because such a delegation of powers was of a “vast nature.”33 The Judicial

Committee of the Privy Council rejected this contention, reiterating that it was “unnecessary to try to enumerate the innumerable occasions in which Legislatures both Provincial and Dominion and Imperial have entrusted various persons and bodies with similar powers to those contained in

[the] Act.”34 Interestingly, the Judicial Committee of the Privy Council endorsed the reasoning of the trial judge in the matter.35 That reasoning supported the idea that the respective legislatures 2020 CanLIIDocs 693 are sovereign within their own spheres, but that there is a principle of non-abdication, acting as a limit on the power that Parliament can delegate.36

This principle is further elaborated in the case of In Re George Edwin Gray,37 “the leading judgment on the constitutionality of delegation,”38 which also contains the first mention of “abdication” as a doctrinal limit on delegation. In that case, Parliament delegated to the

32 Ibid at 87.

33 Sga’nism Sim’augit (Chief Mountain) v Canada (Attorney General), 2013 BCCA 49 at

para 91, 359 DLR (4th) 231.

34 Shannon, supra note 31 at 87.

35 Ibid: “Martin CJBC appears to have disposed of this objection very satisfactorily in his

judgment on the reference, and their Lordships find no occasion to add to what he there

said.”

36 Willis, supra note 16 at 260.

37 Gray, supra note 13.

38 Neudorf, supra note 1 at 534.

Governor in Council (during World War I) the “power to do and authorize such acts and things and to make from time to time such orders and regulations as he may, by reason of the existence of real or apprehended war, deem necessary or advisable for the security, defence, peace, order and welfare of Canada.”39 Clearly, this was a broad delegated legislative power. In fact, John

Willis says that the statute had the effect of “transferr[ing] to the executive the whole legislative authority of the Dominion Parliament for the duration of the War.”40 The Court upheld the delegation, with Chief Justice Fitzpatrick providing the legal position that “[t]he practice of 2020 CanLIIDocs 693 authorizing administrative bodies to make regulations to carry out the object of an Act…is well known and its legality is unquestioned.”41 This line is a key takeaway from Gray and is consistent with cases like Hodge.

Fitzpatrick C.J.C. further noted that “Parliament cannot, indeed, abdicate its functions, but within reasonable limits at any rate it can delegate its powers to the executive government.”42

Justice Duff (as he then was) said in concurrence that, while the delegation was limited in scope,

“there is no attempt to substitute the executive for parliament in the sense of disturbing the existing balance of constitutional authority.”43 Justice Anglin added that “[a] complete abdication by Parliament of its legislative functions is something so inconceivable that the constitutionality

39 Gray, supra note 13 at 161.

40 Willis, supra note 16 at 256.

41 Gray, supra note 13 at 156.

42 Ibid at 157.

43 Ibid at 170.

of an attempt to do anything of the kind need not be considered.”44 Anglin J. recognized that there would be a limitation of “non-abdication,” but that the limit would appear obvious. The main points from Gray are that “there is no constitutional objection to…extremely extensive delegation” and that the Court recognized a principle of non-abdication, albeit narrowly defined.45

Further refinements to this nascent principle of non-abdication came in the Reference as to the Validity of the Regulations in relation to Chemicals enacted by Order in Council and of an 2020 CanLIIDocs 693 Order of the Controller of Chemicals made pursuant thereto46 where the Supreme Court of

Canada again upheld a broad delegation of power. Chief Justice Duff also noted, most relevantly for the principle of responsible government, that while the delegation was broad, “the final responsibility for the acts of the Executive rest[ed] upon Parliament. Parliament abandon[ed] none of its powers, none of its control over the Executive, legal or constitutional.”47 Duff C.J.C. appeared to reject the possibility of a constitutional issue as to the scope of the delegation, saying that while there is “some risk of abuse when wide powers are committed in general terms to any body of men,”48 Parliament had mechanisms of ex post control over the delegation. Justice

Rinfret similarly noted that “Parliament [had] not abdicated its general legislative powers...it

[had] indicated no intention of abandoning control and [had] made no abandonment of

44 Ibid at 176.

45 Willis, supra note 16 at 257.

46 Chemicals Reference, supra note 14 at 12.

47 Ibid.

48 Ibid.

control.”49 This was because the Cabinet “remain[ed] responsible directly to Parliament and depend[ed] upon the will of Parliament for the continuance of its official existence.”50 For

Rinfret J., the focus was retention of parliamentary power: “[Parliament] retains its powers intact, and can, whenever it pleasures, destroy the agency it has created and set up another, or take the matter directly into its own hands.”51 Justice Kerwin gave similar reasons, saying that the “remedy lies in [Parliament’s] own hands” should it feel that “too great” a power has been conferred.52 In other words, courts have no say on the scope of delegated power. 2020 CanLIIDocs 693 The Chemicals Reference adds to the principles from Re Initiative and Referendum Act and Gray by explaining what abdication is not. For both Duff C.J.C. and Rinfret J., so long as some responsibility can be linked back to Parliament, it is within Parliament’s domain to delegate its power away in as broad a scope as it wishes. With the executive, this link occurs with political accountability. So long as that link exists, Parliament does not abdicate its control because, as Rinfret J. noted, the Cabinet “remains responsible directly to Parliament.”53 In this way, as per Kerwin J., the remedy ultimately lies with Parliament.

Delegation was also considered in Re: Authority of Parliament in relation to the Upper

House.54 In that case, the Supreme Court of Canada contemplated whether “the Parliament of

49 Ibid at 18.

50 Ibid.

51 Ibid at 26 [emphasis added], citing Hodge, supra note 23.

52 Chemicals Reference, supra note 14 at 30.

53 Ibid at 18.

54 [1980] 1 SCR 54, 1979 CanLII 169 [Upper House Reference cited to SCR].

Canada [had] legislative authority to abolish the Senate.”55 The Court held that such a change would “alter the structure of the federal Parliament to which the federal power to legislate is entrusted under s. 91 of the [Constitution].”56 It would have done so by transferring “the legislative powers enumerated in s. 91 to some body or bodies other than those specifically designated in it.”57 Relying on Re Initiative and Referendum Act, the Supreme Court of Canada held that “[t]he elimination of the Senate would go much further in that it would involve a transfer by Parliament of all of its legislative powers to a new legislative body of which the 2020 CanLIIDocs 693 Senate would not be a member.”58 The Upper House Reference supports the principle set out in

Re Initiative and Referendum Act: Parliament cannot outrightly transfer its powers to a new legislative body not contemplated by the constitutional text.

Taking these cases together, one can draw out some tentative principles. The cases recognize a general principle of non-abdication. That principle seems connected to the idea of responsible government, central to the Westminster system of government. Notably, the cases are largely silent on the precise connection between the rule of non-abdication and what is

55 Ibid at 59.

56 Ibid at 66.

57 Ibid at 72. Section 91 specifically designates legislative bodies as follows: “It shall be

lawful for the Queen, by and with the Advice and Consent of the Senate and House of

Commons, to make Laws for the Peace, Order, and good Government in Canada”

(Constitution Act, 1867 (UK), 30 & 31 Vict, c 3, s 91, reprinted in RSC 1985, Appendix

II, No 5).

58 Upper House Reference, supra note 54 at 72.

required of Parliament when it delegates its power. On first impression, there appears to be an idea or requirement that Parliament must control the delegation as it cannot create a new law- making body not contemplated by the Constitution. However, there is no impropriety in delegation as such, nor on the breadth of power delegated.

B. AMERICAN DELEGATION RULES

As a point of contrast, the requirement of control does not arise in the American non-delegation 2020 CanLIIDocs 693 doctrine. Rather than establishing a norm of control, the American doctrine is concerned with limiting the scope of delegation as a function of the strict separation of powers in the United

States. This distinction sheds light on the difference between scope versus recipient restrictions on delegation and the argument that a scope restriction might apply in a Westminster system of parliamentary democracy.

The strict formulation of the American non-delegation doctrine is simple: Congress may not delegate its legislative power to the executive.59 The historical development of this doctrine, outlined in cases from the Supreme Court of the United States,60 reveals three justifications.

First, the Supreme Court of the United States casts the doctrine as a function of the separation of powers.61 Second, the text of the Constitution of the United States arguably supports a strict non- delegation doctrine. Article I of the Constitution of the United States, as noted above, clearly

59 Mistretta v United States, 488 US 361 at 371–72 (1989) [Mistretta].

60 See JW Hampton Jr & Co v United States, 276 US 394 (1928); ALA Schechter Poultry

Corp. v United States, 295 US 495 (1935); Yakus v United States, 321 US 414 (1944).

61 See Mistretta, supra note 59 at 371–72.

states that all legislative power rests in the Congress of the United States.62 This argument implies that Congress cannot delegate the power to make laws to someone else. Finally, there is the bicameralism and presentment concern. Article I, § 7 of the Constitution of the United States provides a “finely wrought and exhaustively considered…procedure” for making laws.63 When agencies “make law,” they do not abide by this procedure.

These three arguments, each rooted in a strong-form version of the separation of powers particular to the American context, are marshalled to justify an absolute prohibition on the 2020 CanLIIDocs 693 delegation of legislative powers. This view was endorsed by the Supreme Court of the United

States in Whitman v. American Trucking Associations, Inc.,64 where Justice Scalia, writing for the majority, concluded that there is no amount of legislative power that can be delegated.65

While the justifications for the doctrine rest in textualism and the separation of powers, the form of the doctrine is far more permissive, focusing on policing the boundaries of the scope of delegated power.66 The doctrine provides that Congress can delegate its power so long as it

62 US Const art I, §7.

63 Immigration and Naturalization Service v Chadha, 462 US 919 at 951 (1983).

64 531 US 457 (2001).

65 Ibid at 472.

66 Keith E Whittington & Jason Iuliano, “The Myth of the Nondelegation Doctrine” (2017)

165:2 U Pa L Rev 379 at 380: “Nondelegation doctrine cases follow a predictable pattern.

Every few years, a court of appeals invokes the doctrine to strike down a federal statute.

The Supreme Court inevitably grants certiorari and overturns the appellate decision,

holding that the statute is a constitutional delegation of legislative authority.”

provides an “intelligible principle” to guide the delegation.67 The basic idea of the doctrine is that

Congress cannot delegate its legislative power to the executive but can confer discretion on the executive to implement and enforce a law that Congress has already passed,68 thereby providing the executive substantial powers to delineate the law. In this way, the intelligible principle test is a fictional mechanism that converts the delegation of power that could be characterized as law- making power into discretionary “executive” power.69 The courts do not decide whether a particular delegation of legislative power is too great, as no legislative power is formally 2020 CanLIIDocs 693 delegated; instead, the courts determine whether an intelligible principle sufficiently sets the scope of discretionary authority given to the executive.70

The ex ante non-delegation doctrine is fundamentally concerned with the scope of the delegation and the prevention of discretionary or arbitrary administrative authority.71 In this

67 For a recent description, see Gundy v United States, 588 US (2019) (slip op.) at 2, 5

[Gundy]. For alternative versions of how the Supreme Court of the United States has

instantiated a non-delegation rule, see Cass R Sunstein, “Nondelegation Canons” (2000)

67:2 U Chicago L Rev 315; Industrial Union Department v American Petroleum

Institute, 448 US 607 (1980).

68 See Gundy, supra note 67 at 2, 5.

69 See ibid. See also City of Arlington, Texas v Federal Communications Commission, 569

US 290 at n 4 (2013): “Agencies make rules…[that] take ‘legislative’…forms, but they

are exercises of…the ‘executive Power.’”

70 See Mistretta, supra note 59 at 415.

71 Kenneth Culp Davis, “A New Approach to Delegation” (1969) 36:4 U Chicago L Rev

understanding of non-delegation, the intelligible principle narrows the scope of the delegation by providing guidance to executive actors exercising delegated authority. It sets the appropriate standard for how executives should exercise their discretion in accordance with the separation of powers. As Bogdan Iancu argues, this is a particular type of non-delegation limit concerned with the vagueness of a particular delegated power and “the possibility of abuse in the absence of a clear posited rule.”72 A non-delegation limit attaches here because of the possibility of

“unfettered executive, administrative, or judicial discretion.”73 2020 CanLIIDocs 693 Most notably, this focus on the scope of delegated power is particular to the American context of separated powers. As mentioned above, the goal is to ensure that, as much as possible, legislative power, vested in the Congress of the United States by Article I, § 1,74 is not transferred to executive hands. This grant of power limits Congress. As we shall see, there is no such applicable limitation in the Canadian system of parliamentary sovereignty.

III. NON-ABDICATION AS A RESTRICTION ON RECIPIENT RATHER THAN

SCOPE

Older Canadian cases seem to suggest two focal points of analysis that differ from the American non-delegation doctrine. The first focus is the recipient of the delegated power. In Re Initiative

713.

72 Iancu, supra note 12 at 4.

73 Ibid.

74 US Const art I, § 1.

and Referendum Act75 and Upper House Reference,76 the recipient of the power was a body unknown to the constitutional tradition. In Gray,77 on the other hand, the recipient was the

Cabinet. The second focus is the scope of the delegated power. For example, Hodge speaks of the power being an authority “ancillary” to legislation.78 This suggests that constitutionally permissible delegations must be limited, as they are in the American non-delegation doctrine.

This fault line between scope and recipient can inform the meaning of what the court addressed in the non-abdication cases. For Willis, for example, the statements of non-abdication 2020 CanLIIDocs 693 import the American non-delegation doctrine into Canadian law.79 Similarly, Lorne Neudorf states that “[c]ourts should be much slower in accepting the delegation of sweeping powers to the executive,”80 advocating an approach that requires “generic words” to be capable of

“intelligent qualification by the text, context, or purpose of the statute.”81 (Alyn) James Johnson suggests that it is the “constitutional role of the courts to ensure that delegations contain adequate substance to guide and control the resulting executive decisions.”82 For all of these authors, it appears that a scope-based approach, similar to the American doctrine, applies in Canadian law.

Despite these arguments, it does not appear that the scope-based non-delegation limit has

75 Supra note 13.

76 Supra note 54.

77 Supra note 13.

78 Supra note 23 at 12.

79 See generally Willis, supra note 16 at 254.

80 Neudorf, supra note 1 at 557.

81 Ibid at 560.

82 Johnson, supra note 9 at 832.

application in the Canadian context, in part because of the history of parliamentary sovereignty.

Parliamentary sovereignty, while subject to constitutional limits not relevant to this discussion, means that Parliament can delegate any scope of power it sees fit to actors without judicial approbation. Non-abdication, then, as a judicial doctrine, must attach to the recipient of delegated power, restricting the class of persons and bodies able to accept delegated power. This is in stark contrast to the American system of separated powers. It means that arguments like those of Willis, Neudorf, and Johnson have no place in the context of a Canadian system of 2020 CanLIIDocs 693 parliamentary sovereignty.

This section is divided into three parts. In Part A, I show why parliamentary sovereignty and responsible government remain relevant principles in Canadian constitutional law. This is important because if Parliament is sovereign within constitutional limits, then there is no warrant to impose any limits on its ability to delegate, specifically on the scope of delegation like in the

U.S. In Part B, I demonstrate why this conclusion remains true, even in the era of the Canadian

Charter of Rights and Freedoms83 and the recognition of unwritten constitutional principles.

Finally, in Part C, I set out a rule of non-abdication that is consistent with the status of Canada’s constitutional principles and contrast it with the scope-based American non-delegation doctrine.

A. THE CONTINUED RELEVANCE OF PARLIAMENTARY SOVEREIGNTY IN

CANADA

The Canadian Constitution comprises written instruments and unwritten conventions and

83 Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK),

1982, c 11 [Charter].

principles. In terms of written instruments, the Canadian Constitution principally consists of the

Constitution Act, 1867, which divides powers between the provinces and the federal government, and the Constitution Act, 1982,84 which includes the Charter that catalogues certain protected rights and freedoms. An organizing principle of these constitutional documents, and the unwritten conventions and principles that govern them, is that of parliamentary sovereignty. That principle states that, within constitutional limits, Parliament can make whatever laws it wishes, including laws that delegate broad power to administrative actors. Restrictions on scope, then, 2020 CanLIIDocs 693 have no force in this context.

The starting point is the preamble to the Constitution Act, 1867, which states that Canada is to have a constitution “similar in Principle to that of the United Kingdom.”85 In a judicial reference regarding the remuneration of provincial judges,86 the Supreme Court of Canada noted that the preamble to the Constitution, while not itself a source of positive law,87 has “important legal effects”88 through which courts can fill gaps in the Constitution.

84 Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11.

85 Supra note 57.

86 Reference re Remuneration of Judges of the Provincial Court of Prince Edward Island;

Reference re Independence and Impartiality of Judges of the Provincial Court of Prince

Edward Island, [1997] 3 SCR 3, 1997 CanLII 317 [Provincial Judges Reference cited to

SCR].

87 Ibid at para 94.

88 Ibid at para 95.

The preamble sustains the constitutionalization of parliamentary privileges89 and the inference of a representative, parliamentary institution.90 Indeed, the position of Parliament in the

Canadian state is a direct descendant of its British predecessor. While a history of the British

Parliament is beyond the scope of this paper, it is said that the British version of Parliament—the mother of all parliaments—won its sovereignty in the Bill of Rights [1688].91 Issued in response to the acts of James II, the English Bill of Rights was the beginning of a system of parliamentary sovereignty, because the Crown could no longer “suspend” or “dispense with” the laws without 2020 CanLIIDocs 693 parliamentary approval.92 Indeed, coupled with the Case of Proclamations some 78 years earlier, it became the case that the Crown could not purport to legislate without parliamentary approval—and these principles remain relevant to English law today.93

89 Ibid at para 101.

90 Ibid at para 100. See also Saumur v City of Quebec, [1953] 2 SCR 299 at 330, 1953

CanLII 3.

91 (UK) 1 Will & Mar, sess 2 c 2 [Bill of Rights].

92 See e.g. David V Williams, “To Remind People of the Bill of Rights 1688” (1977) 3:3

Monash UL Rev 243 at 243: Provisions of the Bill of Rights “established once and for all

that regal authority may not be exercised in an arbitrary manner without regard for the

laws laid down by Parliament.” Other statutes between the years 1689 and 1707 were

also particularly important in this respect: Act of Settlement 1700 (Eng), 12 & 13 Will III,

c 2; Claim of Right Act 1689 (Scot), 1689, c 28; Union with Scotland Act 1706 (Eng), 6

Ann, c 11; Union with England Act 1707 (Scot), 1707, c 7.

93 R v Miller, [2017] UKSC 5 at para 177: “I entirely accept the importance in our

In connection with this, one of the innovations of the Bill of Rights was that Parliament could empower and control the Crown as it saw fit,94 and Parliament could conceivably delegate any legislative power to the Crown. This meant that, in some part, the Crown could only be enabled by Parliament. It has been said that “each successive delegation of legislative power has been a fresh recognition of that supremacy,” each a “victor[y] at the expense of the Crown,”95 in which the Crown gives up the pretension to legislate by itself.96

But this is not a matter of vintage—in light of these principles, the Supreme Court of 2020 CanLIIDocs 693 Canada has recognized that it still operates under the “grundnorm” of parliamentary sovereignty.97 There are many ways that this is true, but consider first the express acknowledgement of “democracy” as an unwritten principle of the Canadian Constitution in

Reference re Secession of Quebec.98 As noted above, the Canadian Constitution is a mixture of

constitutional law of the principle of Parliamentary sovereignty over our domestic law,

established in the Case of Proclamations.”

94 Walter Bagehot, The English Constitution, 1st ed (London: Chapman & Hall, 1867) at

314: “The rule of Parliament was definitely established in 1688.”

95 CT Carr, Delegated Legislation: Three Lectures (London: Cambridge University Press,

1921) at 48.

96 Ibid at 52.

97 Canada (Auditor General) v Canada (Minister of Energy, Mines and Resources), [1989]

2 SCR 49 at 103, 1989 CanLII 73 [Minister of Energy, Mines and Resources].

98 [1998] 2 SCR 217, 1998 CanLII 793 [Secession Reference cited to CanLII].

written instruments and unwritten conventions and principles.99 In the Secession Reference, the

Supreme Court of Canada set out that democracy, as a fundamental unwritten principle of the

Constitution, was expressly connected to the idea of a supreme Parliament as established in the

1688 Revolution.100 Specifically, the principles of the Bill of Rights and the unwritten principles of the Canadian Constitution are part of its “internal architecture and structure,”101 which define

“the delineation of spheres of jurisdiction…and the role of our political institutions.”102 In

Canada, the principle of democracy maintains the importance of parliamentary sovereignty. 2020 CanLIIDocs 693 These principles hold true in the present day, contrary to some modern commentary.103

Recently, in the Pan-Canadian Securities Reference, the Court discussed the nature of parliamentary sovereignty in Canada. While noting that the written Canadian Constitution

“qualified the basic Diceyan rule” of sovereignty, it also noted that parliamentary sovereignty was “an equally important feature of Canadian law.”104 It went even further, suggesting that

“the…principle of parliamentary sovereignty remains foundational to the structure of the

Canadian state: aside from these constitutional limits, the legislative branch of government

99 Re: Resolution to Amend the Constitution, [1981] 1 SCR 753 at 877–80, 1981 CanLII 25.

100 Secession Reference, supra note 98 at para 63.

101 Régimbald & Newman, supra note 10 at 67.

102 Secession Reference, supra note 98 at para 52.

103 In Part III-B, below, I explore whether any modern limitations on parliamentary

sovereignty affect the issue of delegation.

104 Pan-Canadian Securities Reference, supra note 21 at para 56.

remains supreme over both the judiciary and the executive.”105

Parliamentary sovereignty in the purest sense is not the only relevant principle to the problem of delegation. The sovereignty doctrine not only empowers Parliament as an institution but also places correlative duties on the executive in terms of responsibility. The executive must be accountable for the use of delegated powers. As recently noted by the Supreme Court of the

United Kingdom in assessing the lawfulness of a parliamentary prorogation,106 “the policies of the executive are subjected to considerations by the representatives of the electorate, the 2020 CanLIIDocs 693 executive is required to report, explain and defend its actions, and citizens are protected from the arbitrary exercise of power.”107 In other words, the executive cannot undermine powers delegated to it. It must execute them faithfully and account to the legislature for their exercise.

Accountability is instantiated through the convention of responsible government.108 The term “responsible government” did not enter the Canadian lexicon until the nineteenth century,109 but it remains the main method by which Parliament exercises control over delegated

105 Ibid at para 58 [emphasis added].

106 R (on the application of Miller) v Prime Minister; Cherry and others v Advocate General

for Scotland, [2019] UKSC 41, [2019] 4 All ER 299.

107 Ibid at para 46.

108 Conventions are rules of political practice that, while not judicially enforceable, are

considered binding by political actors in the system.

109 See generally Sir CP Lucas, Lord Durham’s Report, vol 2 (Oxford: Clarendon Press,

1912). Lord Durham was sent by the English Crown in 1838 to investigate the causes of

rebellion in British North America. He recommended a system of responsible

actors. In the British and Canadian contexts, the principle generally means that individual ministers are responsible to Parliament for the work of their departments, and the Cabinet is collectively responsible for government policy. Responsible government enshrines the principle that the executive arm of government is responsible to an elected assembly—and thus the people.

The goal is to subordinate the Crown to Parliament to promote political control and compliance.

The convention of responsible government is the way by which this subordination is operationalized, and by which parliamentary sovereignty is constitutionally empowered. 2020 CanLIIDocs 693 Together, these form the “mainspring of the Constitution.”110 It has been said that “[t]he whole life of English politics is the action and reaction between the Ministry and the Parliament.”111

The entire principle is one of balance: Parliament simultaneously controls and empowers the executive.112 Failure to abide by this balance would “create a new theory of government.”113

Indeed, the Supreme Court of Canada has endorsed the basic principle of responsible government as a key part of Canada’s constitutional arrangements. The Court has noted that “the tenets of responsible government…underlie the discharge of legislative authority under the

British North America Act,”114 while noting in another decision that the preamble sustains the

government for Canada.

110 Ibid.

111 MJC Vile, Constitutionalism and the Separation of Powers (Indianapolis: Liberty Fund,

1967) at 250.

112 Ibid at 241, 242.

113 Ibid at 258.

114 Curr v The Queen, [1972] SCR 889 at 899, 1972 CanLII 15.

principle of responsible government as at least a background organizing premise of the Canadian state.115 More broadly, the Court has spoken about the general principle of the executive’s subordination to Parliament,116 holding in the Secession Reference that the Constitution mandates responsible government: “The Constitution mandates government by democratic legislatures, and an executive accountable to them.”117

The core idea, when it comes to delegation, is that when the executive uses delegated legislative power, Parliament continues to control the exercise of these powers. As noted in the 2020 CanLIIDocs 693 Chemicals Reference and Gray,118 there is always a natural mechanism of control over the exercise of these powers. This mechanism of control is always required as a constitutional matter and is effective in ensuring that delegated actors are subject to the principle of democracy.

The takeaway from this section is twofold. First, parliamentary sovereignty and responsible government are not dead in Canada, even under a general principle of constitutional sovereignty. As recognized by the Supreme Court of Canada, parliamentary sovereignty undergirds the framework of the Canadian state.119 Second, there is a specific reference to the idea of sovereignty as relational. In the Pan-Canadian Securities Reference, the Supreme Court of Canada is careful to elucidate the constitutional limits on Parliament but is equally careful to note that Parliament is supreme in reference to the executive and the judiciary.120 This implies a

115 See Ontario (Attorney General) v OPSEU, [1987] 2 SCR 2, 1987 CanLII 71 at para 85.

116 Johnson, supra note 9 at 850.

117 Secession Reference, supra note 98 at para 68. 118 See Part II-A, above.

119 See Minister of Energy, Mines and Resources, supra note 97 and accompanying text.

120 See supra notes 103–105 and accompanying text.

subordination in which Parliament can control or empower the executive as it chooses within constitutional limits.

A consequence of parliamentary sovereignty is that, absent any constitutional limitations, a court cannot limit the scope of delegated power. This is because Parliament is sovereign with reference to both the judiciary and the executive. It can delegate whatever power it wishes, and that power is only limited by the restrictions evident in the non-abdication rule itself. This is in contrast to the American system of separated powers in which explicit textual limitations ensure 2020 CanLIIDocs 693 that Congress cannot delegate legislative power.

B. PUTATIVE LIMITS ON PARLIAMENTARY SOVEREIGNTY

Parliamentary sovereignty permits Parliament to pass “any law whatever”121 and therefore undermines an argument in favour of limiting the scope of legal power delegated to the executive. It is true, however, that the Canadian instantiation of parliamentary sovereignty is not as pure as the British version.122 Unlike the United Kingdom, Canada has a written Constitution that imposes limits on the principle of parliamentary sovereignty as it pertains to lateral delegation. Indeed, the Supreme Court of Canada has stated, on many occasions, that Canada has developed into a system of “constitutional supremacy” over “parliamentary sovereignty.”123 This

121 Dicey, supra note 11 at 38.

122 J Noel Lyon, “The Central Fallacy of Canadian Constitutional Law” (1976) 22:1 McGill

LJ 40 at 43: There is an “unexamined assumption that Canadian legislatures enjoy a

supremacy of the same quality as that of the Parliament of the United Kingdom.”

123 Secession Reference, supra note 98 at para 72; Pan-Canadian Securities Reference, supra

development is attributable to the adoption of the Constitution Act, 1982, which entrenched a judicially enforceable catalogue of rights.124 For those who argue for ex ante limitations on delegated power, this is significant because it limits the scope of parliamentary power.

Additionally, one might argue that other unwritten principles of Canadian constitutional law, including democracy and the rule of law, might attach limits on the scope of delegated power.125

In this section, I review these two potential reasons for restricting parliamentary sovereignty.

2020 CanLIIDocs 693 1. Charter Limitations

For the Constitution to impose restrictions on delegation, it must impose restrictions on parliamentary sovereignty on the specific legislative power exercised by Parliament when it delegates to the executive. In a recent article, Neudorf advances two ways in which the

Constitution could restrict the delegation power: through the general interpretive approach of the

“living tree,” and through a muscular application of the void-for-vagueness doctrine. However, both arguments are left wanting.

First, Neudorf posits that Canada’s so-called “leading” constitutional interpretation doctrine, the “living tree” interpretation, casts doubt on the approach of the Supreme Court of

Canada in the leading delegation cases.126 Neudorf says that these cases reflect a “narrow and

note 21 at para 58; Reference re Supreme Court Act, ss. 5 and 6, 2014 SCC 21 at para 89,

[2014] 1 SCR 433.

124 There are also limitations on delegated power that are derivative of the distribution of

powers: see NS Interdelegation, supra note 13.

125 See generally Johnson, supra note 9. 126 Neudorf, supra note 1 at 542, 544.

technical reading” of the Constitution.127 Particularly, Neudorf posits that a “purely textual reading” of the Constitution “cuts against the grain of the living tree.”128 Instead, he argues that a living tree approach recognizes the developed “role and purpose of Parliament” in the twenty- first century and the “democratic, representative and accountable” nature of law-making in

Canada.129

Neudorf’s reliance on the living tree approach to interpretation is misplaced for two reasons as it relates to delegation. First, the living tree justification for Neudorf’s non-delegation 2020 CanLIIDocs 693 doctrine is weak. As Neudorf notes, the living tree approach to interpretation purportedly began in Edwards v. Canada (Attorney General).130 Lord Sankey’s comments in the Persons Case have been ascribed almost-mythical status in establishing a rule of constitutional interpretation: “The

British North America Act planted in Canada…a living tree capable of growth and expansion within its natural limits.”131 The idea, relied on by defenders of this sort of progressive interpretation, is that the Constitution of Canada should grow to fit the realities of the period.132

But the intellectual credentials of the living tree doctrine have come under scrutiny.

127 Ibid at 544.

128 Ibid.

129 Ibid at 546.

130 [1929] UKPC 86, [1930] 1 DLR 98 [Persons Case cited to DLR].

131 Ibid at 106–107.

132 See Justice Ian Binnie, “Constitutional Interpretation and Original Intent” (2004) 23

SCLR (2d) 345–82, cited in Grant Huscroft & Ian Brodie, Constitutionalism in the

Charter Era (Markham: LexisNexis, 2004) at 345–48.

Professor Bradley W. Miller’s (as he then was) consideration of the Persons Case demonstrates instead that the living tree metaphor belies the rest of the judgment, in which the Judicial

Committee of the Privy Council arguably engaged with methods of public meaning originalism and the semantic meaning of the constitutional text.133 The living tree approach was not picked up in Canada after the Persons Case, in part because “it is entirely absent from, and antithetical to, the Privy Council’s methodology in the Persons Case.”134 It was only in the Charter era that the Supreme Court of Canada began using the living tree metaphor; and in the last ten years, the 2020 CanLIIDocs 693 Court has sparingly mentioned it.135 It is odd that Neudorf relies on this particular metaphor when its constitutional credentials are being questioned.

On the specific issue of delegation, it is not clear that a living tree approach would favour

Neudorf’s vagueness doctrine or any other ex ante restriction on delegation. Neudorf argues that the previous Supreme Court of Canada cases upholding delegation were the result of a restricted view of the Parliament of Canada, which has now developed into a distinct institution from its

133 Bradley W Miller, “Origin Myth: The Persons Case, the Living Tree, and the New

Originalism” in Grant Huscroft & Bradley W Miller, eds, The Challenge of Originalism:

Theories of Constitutional Interpretation (Cambridge: Cambridge University Press,

2011) 120 at 129–30, 134.

134 Ibid at 137.

135 See Asher Honickman, “Has the Supreme Court Moved Beyond the ‘Living Tree’?” (30

March 2018), online: Advocates for the Rule of Law

supreme-court-moved-beyond-the-living-tree>, archived: .

British antecedent.136 It follows, then, that a living tree approach would consider the nature of

Parliament’s law-making function in the modern Canadian state. But it is equally likely that a living tree approach would weigh in favour of recognizing the important role Parliament plays in the constitutional structure. Read this way, Gray and the other cases recognize the principle that

Canada’s early Parliament was supreme and thus free to delegate law-making power. The living tree approach, therefore, does not necessarily speak to the particular issue of lateral delegation as

Neudorf suggests. 2020 CanLIIDocs 693 Neudorf also posits that the vagueness doctrine, which arises under s. 7 of the Charter,137 can be applied to delegation situations. Specifically, he argues that “when generic words are used in enabling legislation, which are incapable of intelligent qualification by the text, context or purpose of the statute, the court should hold the grant of authority invalid on the basis that it is impermissibly vague.”138

This statement strongly resembles the American ex ante non-delegation doctrine.

Respecting the vagueness doctrine, the Supreme Court of Canada speaks of a requirement that

“[a] law must set an intelligible standard both for the citizens it governs and the officials who must enforce it.”139 Neudorf suggests that the same restriction could apply to the scope of

136 Neudorf, supra note 1 at 544–46.

137 Supra note 83, s 7: “Everyone has the right to life, liberty and security of the person and

the right not to be deprived thereof except in accordance with the principles of

fundamental justice.”

138 Neudorf, supra note 1 at 560.

139 Canadian Foundation for Children, Youth and the Law v Canada (Attorney General),

delegated power.140

The Charter does not, however, regulate the relationships between branches of government as such. This is in direct contrast to the Constitution Act, 1867, and even the Bill of

Rights, both of which regulate the relationships between orders of government and Parliament and the Crown, respectively. Rather, the Charter regulates the relationship between the individual and the state.141 In other words, Neudorf suggests that the vagueness doctrine applies to the legislation delegating power itself, rather than the application of discretion under delegated 2020 CanLIIDocs 693 power. The latter may certainly raise constitutional issues, but the former affects no rights or freedoms.

The limitations of the vagueness doctrine are reflected in the Supreme Court of Canada’s application of it. The doctrine has mostly been applied either to criminal offences enacted by

Parliament or to offences under which a person was charged and raised some constitutional issue.142 In fact, in one of the leading vagueness doctrine cases, the Supreme Court of Canada

2004 SCC 4 at para 16, [2004] 1 SCR 76 [CFCY].

140 Neudorf, supra note 1 at 561.

141 RWDSU v Dolphin Delivery Ltd., [1986] 2 SCR 573, 1986 CanLII 5 at paras 26–41.

142 See R v Nova Scotia Pharmaceutical Society, [1992] 2 SCR 606, 1992 CanLII 72 (law

pertaining to conspiracy to lessen competition, under which the appellants were charged);

R v Morgentaler, [1988] 1 SCR 30, 1988 CanLII 90 (Criminal Code (RSC, 1985, c C-46)

provision criminalizing abortion and providing restrictions on abortion, under which a

doctor who performed abortions was charged); Irwin Toy Ltd. v Quebec (Attorney

General), [1989] 1 SCR 927, 1989 CanLII 87 (Quebec provincial government brought

noted that “[a] vague law prevents the citizen from realizing when he or she is entering an area of risk for criminal sanction.”143 This makes sense. As criminal laws and the enforcement of those laws could infringe rights and freedoms by their very nature, vagueness could be a major constitutional concern. The relationship regulated is between the individual and the state, and the constitutional attack is not aimed at Parliament’s delegatory power. For that reason, the vagueness doctrine is likely inapplicable to limit the scope of delegated power.

More importantly, the void for vagueness doctrine requires a notoriously high standard 2020 CanLIIDocs 693 and has been judicially recognized in only a handful of cases—none of which involved delegated legislative power.144 Even if the void for vagueness doctrine is applicable and relevant to delegated legislation, it would not be a meaningful restriction on delegated power in the way

Neudorf suggests.

2. Unwritten Principles

Another potential basis for imposing restrictions on delegated power is through unwritten constitutional principles,145 two of which are particularly relevant: democracy and the rule of

contravention charges under consumer protection law against a respondent who

broadcasted advertising messages); R v Keegstra, [1990] 3 SCR 697, 1990 CanLII 24 (an

individual was charged under a hate speech law).

143 CFCY, supra note 139 at para 16 [emphasis added].

144 See note 142 and accompanying text.

145 See Secession Reference, supra note 98; British Columbia v Imperial Tobacco Canada

Ltd., 2005 SCC 49, [2005] 2 SCR 473 [Imperial Tobacco];

law. However, unwritten principles are generally ill-fitted to create a non-delegation doctrine and, in each case, there are reasons why they cannot be invoked to create such a doctrine.

As noted above, unwritten principles operate in the background of Canada’s constitutional arrangements.146 Indeed, sometimes they can do more: they can “constitute substantive limitations upon government action,”147 and they can give rise to philosophical commitments regarding Canada’s constitutional arrangements.148 However, they have a limited role to play in crafting doctrine: they cannot be used to “dispense with the written text of the 2020 CanLIIDocs 693 Constitution.”149 Speaking specifically of the rule of law, the Supreme Court of Canada has noted that the principle cannot “be used as a basis for invalidating legislation…based on its content.”150 In short, unwritten principles are useful guides for constitutional interpretation; however, the extent of their use and how they should be used (e.g. to fill gaps in the Constitution

146 Secession Reference, supra note 98 at para 50: unwritten principles “infuse our

Constitution and breathe life into it.”

147 Ibid at para 54. See also Provincial Judges Reference, supra note 86 at para 185.

148 Kate Glover Berger, “The Demands of Unwritten Constitutionalism on Institutional

Design” (29 May 2019), online (blog): IACL-AIDC

constitutional-principles/2019/5/25/the-demands-of-unwritten-constitutionalism-on-

institutional-design>, archived: .

149 Secession Reference, supra note 98 at para 53.

150 Imperial Tobacco, supra note 145 at para 59.

or create judicial doctrine) remain open questions.151

With that limitation in mind, I turn first to the argument that the unwritten principle of democracy has the potential to create a non-delegation doctrine on scope. This argument is advanced by Johnson, who contends that a “marginalized legislature delegating un-cabined power to willing executive instrumentalities is incoherent and unprincipled” under existing

Supreme Court of Canada precedent.152 Legislatures, so the argument goes, are primarily fora for discussion and deliberation.153 Legislatures effect democratic aims by providing a site for 2020 CanLIIDocs 693 disagreement on complex issues: a form of conflict resolution.154 But as Johnson contends, proponents of unrestricted delegation “fail to offer a theory of democracy” that takes account of this institutional reality of the legislature.155 Indeed, according to Johnson, delegation without restriction fails to account for the democratic requirement that those subject to the law can recognize “the imprimatur of a legislatively resolved conflict.”156 If this is true, then it need not matter if responsible government provides an accountability link back to the legislature: “it will

151 Jean Leclair, “Canada’s Unfathomable Unwritten Constitutional Principles” (2002) 27

Queen’s LJ 389 at 400; Peter W Hogg, “The Bad Idea of Unwritten Constitutional

Principles: Protecting Judicial Salaries” in Adam Dodek & Lorne Sossin, eds, Judicial

Independence in Context (Toronto: Irwin Law, 2010) at 25.

152 Johnson, supra note 9 at 823.

153 Ibid at 845, 850.

154 Ibid at 844.

155 Ibid at 853.

156 Ibid at 878–79.

take more than a brief reference to the fact of responsible government to justify transferring the function of law-making away from an institution that is specifically structured to formulate democratic will.”157

But this argument presupposes that democracy, as an unwritten Canadian constitutional principle, cannot accommodate an unfettered form of parliamentary sovereignty and responsible government that would permit any scope of delegated power to be transferred. That is, it does not take account of the fact that democracy can equally be represented in the act of delegation as in 2020 CanLIIDocs 693 the formation of law ex ante. Johnson’s argument is incorrect as a matter of positive law and on a functional level. On the positive law side, Johnson adopts a different definition of “democracy” than that used by the Supreme Court of Canada. As noted above, the Court’s definition of democracy is intimately related to the Diceyan conception of parliamentary sovereignty. In the

Secession Reference, the Court said that the evolution of our democratic tradition can be traced back to the Magna Carta of 1215 and before, through the long struggle for parliamentary supremacy which culminated in the English Bill of Rights, the emergence of representative political institutions in the colonial era, the development of responsible government in the 19th century, and eventually, the achievement of Confederation itself in 1867.158 Democracy can mean many things, but the Court’s jurisprudence describes a system based on parliamentary sovereignty with an executive linked to the Parliament through responsible government.

The act of legislative delegation still constitutes legislative action that is due respect under the principle of parliamentary sovereignty as a positive law matter. The Supreme Court of

157 Ibid at 870.

158 Secession Reference, supra note 98 at para 62.

Canada, as noted above, has described a legislative act (for Charter purposes) as merely requiring “a norm or standard of general application that has been enacted by a government entity pursuant to rule-making authority.”159 In GVTA, the Court expressly noted that:

A rule-making authority will exist if Parliament or a provincial legislature has

delegated power to the government entity for the specific purpose of enacting

binding rules of general application which establish the rights and obligations of

the individuals to whom they apply…[s]o long as the enabling legislation allows 2020 CanLIIDocs 693 the entity to adopt binding rules, as so long as the rules establish rights and

obligations of general rather than specific application and are sufficiently

accessible and precise, they will qualify as “law.”160

Johnson’s thesis must contend with this positive law statement by the Supreme Court of Canada.

When Parliament delegates power, it sets a norm of general application for the executive to implement. The limits of delegated power—the extent to which a delegate can promulgate rules, and the sort of rules that can be promulgated—depend on the scope of the enabling provision.

Enabling provisions are interpreted, and the scope of delegation is limited and defined, by the objectives and purposes of the provision and its “general nature, scheme, and history.”161 There is no reason to treat any of this as less than a legislative act according to the Supreme Court of

Canada’s statement of what constitutes legislation—indeed, “delegation cannot impeach the

159 GVTA, supra note 3 at para 63.

160 Ibid at para 64 [citation omitted].

161 See Keyes, supra note 1 at 90, 289.

assertion of power.”162

While Johnson attempts to distinguish the act of delegation from the act of formulating policy ex ante,163 this seems to be a distinction without a major difference: both are valid legislative acts.164 Johnson excludes the possibility that there could be substantive deliberation and disagreement by a legislative body over whether and what to delegate in the first place, but this is a mistake. The act of delegation can involve as much discussion and deliberation as a substantive policy question. A legislature may choose to delegate for many reasons, as explored 2020 CanLIIDocs 693 in the political science literature.165 For example, a legislature may decide that a particular question is one that is best solved by expert administrators; this is an empirical judgment about the dimensions of a particular problem.166 Legislatures may also delegate because the costs of

“making law” are too high for political reasons, for example.167 The reasons delegation occurs

162 Reference re Agricultural Products Marketing Association, [1978] 2 SCR 1198, 1978

CanLII 10.

163 Ibid at 846.

164 Ibid at 847. See Edward L Rubin, “Law and Legislation in the Administrative State”

(1989) 89:3 Colum L Rev 369 at 389.

165 Jennifer Nou, “Subdelegating Powers” (2017) 117 Colum L Rev 473 at 475.

166 Jonathan Bendor & Adam Meirowitz, “Spatial Models of Delegation” (2004) 98:2

American Political Science Rev 293 at 294.

167 David Epstein & Sharyn O’Halloran, Delegating Powers: A Transaction Cost Politics

Approach to Policy Making Under Separate Powers (New York: Cambridge University

Press, 1999) at 7–9.

could involve complex questions of policy and legislative judgment, and while there may be a difference between a delegative legislative act and a “pure” legislative act, that difference is not particularly germane to rendering one legislative act more legitimate than another.

Johnson does not stop here. He would also impose rule of law constraints on the ability of legislatures to delegate. Johnson argues that leaving the scope of delegation solely in the hands of legislatures “denies the courts the ability to control arbitrary state conduct at its source.”168

Though, as noted above, the Court in Imperial Tobacco held that the rule of law could not be 2020 CanLIIDocs 693 used to strike the content of legislation, Johnson contends that widespread delegation effaces the

“ordered framework of power relationships”169 that make up Canada’s constitutional framework:

Filling the law with content is a legislative prerogative and responsibility. If the

legislature seeks to transfer this task, an external check is required. Legislation

that alters the fundamental relationship of the central branches of government is

subject to judicial review.170

This line of argument does have some limited support in the case law. For example, the

Provincial Judges Reference could stand for the proposition that the unwritten principle of judicial independence mandates a “special process” to decide judges’ salaries.171 The holding implies that the judiciary must be able to enforce unwritten principles. In Mackin v. New

168 Johnson, supra note 9 at 841.

169 Ibid.

170 Ibid.

171 Provincial Judges Reference, supra note 86 at para 133.

Brunswick (Minister of Finance),172 the Court expanded on the Provincial Judges Reference by holding that the special process at hand must “depoliticize” the relationships between the judiciary and the executive and legislative branches and that such depoliticization is “a structural requirement of the Canadian Constitution resulting from the separation of powers and the rule of law.”173 And in Babcock v. Canada (Attorney General),174 the Court held that the legislature is competent to enact any law “as long as it does not fundamentally alter or interfere with the relationship between the courts and the other branches of government.”175 2020 CanLIIDocs 693 But there are problems with using these cases to argue that the rule of law imposes a distinct non-delegation doctrine on scope in Canada. First, the “institutional relationships” argument implies that delegation somehow upsets those relationships. However, this is not necessarily the case—indeed, delegation may assist the branches of government in performing their intended functions. If regulations are the “life blood of the administrative state,” then surely laws made by the executive assist the legislature in making law in the first place. For that functional reason, more would need to be shown to make the point that delegation necessarily upsets institutional relationships.

Even as a positive law matter, it is difficult to extract a non-delegation doctrine on scope from the cases cited by Johnson for two reasons. First, the cases are simply unclear on the conditions under which unwritten principles could be used to restrict the content or scope of

172 2002 SCC 13, [2002] 1 SCR 405.

173 Ibid at para 69.

174 2002 SCC 57, [2002] 3 SCR 3 [Babcock].

175 Ibid at para 58.

legislative acts. While Babcock alludes to the “institutional relationships” argument, this is a thin reed on which to rest a rather sophisticated non-delegation doctrine. Second, and more importantly, Babcock notes that the use of unwritten principles “must be balanced against the principle of Parliamentary sovereignty.”176 This makes sense: since unwritten principles are not necessarily reflected in the text of the Constitution, they must be used carefully in striking duly- enacted parliamentary legislation. For that reason, it is unclear whether the institutional relationships argument advanced by Johnson would suffice to ground a non-delegation doctrine 2020 CanLIIDocs 693 on scope. Parliamentary sovereignty is itself a fundamental structural principle of the Canadian

Constitution represented in the unwritten principle of democracy. For that reason, Johnson’s thesis must contend with the fact that the Supreme Court of Canada has largely equated the principle of democracy with the role of Parliament.

Altogether, while the unwritten principles argument finds some limited support in the case law, there is not yet enough to suggest that unwritten principles can ground a non-delegation doctrine on scope. This is because the principle of parliamentary sovereignty plays a role in defining Parliament’s powers when it comes to delegation. Particularly, it is unclear how or why the unwritten principles jurisprudence should overturn the statements in Gray and the Chemicals

Reference about the importance of parliamentary sovereignty and responsible government.

C. THE DOCTRINE

Having laid out the relevance of parliamentary sovereignty and responsible government and the inapplicability of particular doctrines that might impact the scope of delegation, I now turn to laying out a defensible construction of a non-abdication rule. That rule attaches to the recipient

176 Ibid at para 55.

of delegated power, rather than to the scope of power delegated. To this extent, it has resonance in the context of independent agencies, which may be intentionally separated from parliamentary control and beyond the strict scope of responsible government.

Assuming that the Supreme Court of Canada’s statements about non-abdication have legal weight and that a rule based on the American non-delegation doctrine would sit uneasily in

Canada, the non-abdication rule would mean that Parliament and ministers must have the opportunity to exercise control over the legislative actions of delegates. In Gray and the 2020 CanLIIDocs 693 Chemicals Reference, this form of control was obvious: the convention of responsible government under which ministers are politically accountable to Parliament. But the calculus changes when it comes to independent agencies with varying relationships to the political executive and for which ministers are not strictly accountable.177 This is shown by the paucity of cases in which the main issue was a delegation to an independent agency rather than a Gray-type delegation to the executive.

To begin the discussion, one can frame the existing principles recognized in the previous sections as follows:

1) Parliament can delegate its law-making powers, subject to the proviso that

it does not “abdicate” its power and establish a new law-making body that

is not subject to Parliamentary control.

2) Abdication occurs when Parliament withdraws control over delegated

actors; specifically, it occurs when delegated actors are not somehow

177 Willis, supra note 16 at 257 speaking of an “implied prohibition against erecting a new

‘legislature.’”

responsible to Parliament.

These principles set out a baseline and establish two basic preconditions: first, there must be a delegation of legislative power; and second, there must be a withdrawal of control. One might say that every time a delegate engages in legislative activity, Parliament should retain the ability, either through ministers or some other means, to approve or veto the particular legislative act— though this is not necessarily the only type of control one can envision.

Before delving into the doctrine, there is a preliminary concern that should be raised. It is 2020 CanLIIDocs 693 true that a pure understanding of parliamentary sovereignty would preclude any court from passing over the contents of parliamentary law. Put differently, if parliamentary sovereignty prevents courts from passing on the scope of delegated power, how can courts construct a non- abdication rule that acts as a fetter against parliamentary sovereignty? There are two main responses to this. First, the non-abdication principle in some form has already been recognized in cases like Re Initiative and Referendum Act and the Upper House Reference. In those cases, it was deployed as a substantive rule of constitutional law. There is therefore precedent for treating the non-abdication rule as a binding and enforceable rule of constitutional law. Second, as I will note below, there are "hooks" in the constitutional text that support the enforceability of this doctrine. This is unlike the restrictions on scope, that, as noted above, do not have a grounding in any constitutional principle. For these reasons, I treat the non-abdication rule as a judicially enforceable rule of constitutional law.

There are textual and general doctrinal reasons why Parliament should retain the ability to control its delegates. Beginning with the text of the Constitution, it is obvious that there is no provision explicitly preventing delegation, and so this “non-abdication” principle is immediately

open to attack as an extra-textual invention.178 However, there are implicit textual and doctrinal signals that demonstrate that the non-abdication cases had merit, and that they constitute a fair interpretation of existing constitutional text and principles. Section 17 of the Constitution Act,

1867, which formally established Canada’s division of powers, says that “[t]here shall be One

Parliament for Canada, consisting of the Queen, an Upper House styled the Senate, and the

House of Commons.”179 Further, other provisions in the Constitution Act, 1867 stipulate that there shall be a “Legislature” in each province. Read harmoniously with s. 17, these provisions 2020 CanLIIDocs 693 seem to indicate that each province should not make another legislature, and Parliament should not make another Parliament. Indeed, the explicit specification of one Parliament implies that there should be no other Parliament that is recognized by the Constitution Act, 1867.180 This has

178 Ibid at 258, speaking of the “total absence from the B.N.A. Act of any express words to

restrict the powers” of legislatures.

179 Supra note 57, s 17.

180 This can be explained through the principle of expressio unius et exclusio alterius: the

inclusion of one thing implies the exclusion of others. The use of the tools of statutory

interpretation in the context of the Constitution Act, 1867 might be controversial. But it

should be recalled that the Constitution Act, 1867 was not viewed as a traditional

constitution in the modern sense; it was apparently understood to be interpreted as an

ordinary statute. See The Bank of Toronto v Lambe, [1887] UKPC 29 at 2: the

Constitution Act, 1867 should be interpreted “by the same methods of construction and

exposition which they apply to other statutes.” See also Asher Honickman, “Unearthing

Canadian Originalism: Reflections on my Conversation with Justice Stratas” (30 January

clear implications for the legal tool of delegation. It means that Parliament cannot delegate its power in a manner that explicitly or implicitly creates another parliament with law-making power.

Similarly, the prefatory clauses in ss. 91 and 92 of the Constitution Act, 1867 speak of an exclusive law-making power held by the federal Parliament and the provinces. If law-making power is held by Parliament and the legislatures, there should be no other body competing for that law-making power, even if by delegation. The exclusivity principle dovetails with the 2020 CanLIIDocs 693 principle of parliamentary sovereignty, which subjects “all the functions of government to the direct control of the legislature.”181 If the government has law-making functions, the legislature should have control over them. This idea is entrenched in the Constitution Act, 1867 and represented by the balance established in the English Bill of Rights in 1688.

The text is also supported by general principles associated with the role of Parliament in the Westminster system, specifically political control and accountability. These concepts concentrate political power so that one can easily trace who is accountable for legislative power.

As noted in the non-abdication cases, when Parliament delegates power straight to a minister, the recipients of the delegated power are strictly accountable for the exercise of that power in the

House of Commons. Everyone—most importantly, the public—knows that a minister with control over a department is politically responsible for that department. Indeed, “without

2019), online: Advocates for the Rule of Law

originalism-reflections-on-my-conversation-with-justice-stratas/#_ftn3>, archived:

181 Vile, supra note 111 at 24.

ministerial responsibility and prime ministerial control of the machinery of government, our

Constitution will not work because the power of the state will not be subject to democratic control.”182 The people control Parliament. Parliament controls the executive through the principle of responsible government. The courts, in some cases, also control the executive.183 All of this establishes the important principle of balance alluded to in the non-abdication cases, by merging control and accountability.184

In Canada, as in the U.S., the rise of the administrative state upset this balance.185 2020 CanLIIDocs 693 Parliament began delegating legislative power to agencies that, while formally part of the executive branch, do not fit easily in the existing structures of government186 and are only

“marginally under the control of [their] political masters.”187 Consequently, a different set of problems arises when Parliament wishes to create an independent agency insulated from political

182 Nicholas d’Ombrain, “Ministerial Responsibility and the Machinery of Government”

(2008) 50:2 Can Public Administration 195 at 197. See also Vile, supra note 111 at 258.

183 Vile, supra note 111 at 254.

184 See Gray, supra note 13 at 170.

185 Sheehy & Feaver, supra note 19 at 197.

186 Ibid at 177: “Agencification was done without attention to the founding principles of

these constitutions, which sought to strike a balance between the exercise of control

powers on the one hand and accountability obligations for the exercise of those powers

on the other. Rather, the agency evolved organically as a solution to specific technical,

economic, and logistical imperatives in distinct historical, political, and legal contexts.”

187 Vile, supra note 111 at 399.

control. In the extreme case of an agency created to make laws with full independence from

Parliament and ministers, Parliament has created an island of power with no connection back to the people. Essentially, this is the Re Initiative and Referendum Act and Upper House Reference problem, and the problem alleged in the Pan-Canadian Securities Reference. In such a case, no one is strictly politically accountable for the exercise of law-making power because ministers are only “answerable”—but not accountable—for the law-making activities of independent agencies.

This divides control and accountability to an “unacceptable degree.”188 Parliament cannot 2020 CanLIIDocs 693 exercise control over ministers in relation to delegated legislative power exercised by agencies because ministers are not responsible for those delegated legislative powers.

For Parliament to countenance such a situation, according to the non-abdication rule, it must maintain the constitutional balance by retaining actual control over those it empowers to create laws. The method of “control and supervision” envisioned by the Supreme Court of

Canada is responsible government: “[F]inal responsibility for the acts of the Executive rests upon

Parliament. Parliament abandons none of its powers, none of its control over the Executive, legal or constitutional.”189 The non-abdication rule could, then, mandate that there be a person politically accountable to Parliament in order to centralize political criticism on one person for the law-making activities of an agency. In other words, there can be no truly independent agency with legislative powers in Canadian law190 because that would establish a new law-making body

188 Sheehy & Feaver, supra note 19 at 197.

189 Chemicals Reference, supra note 14 at 12.

190 HN Janisch, “Independence of Administrative Tribunals in Canada: In Praise of

‘Structural Heretics’” (1988) 8:2 J National Assoc Administrative L Judges 75 at 79–80.

without responsibility for its legislative acts.191

If Parliament delegates power to an independent agency to make regulations, a minister should be held politically accountable for the law-making activities of that agency, just as if the minister was responsible for any law-making activities she conducts on her own. Otherwise,

Parliament should exercise control over the law-making activities of an independent agency directly. If there is a lesser form of scrutiny, such as mere answerability, applied to this delegatory relationship, there is a risk that the lines of authority between the people and the 2020 CanLIIDocs 693 exercise of law-making power could be confused. The worry is that if “the executive is only weakly accountable for the direction of the state by shielding itself from accountability through the use of agencies,”192 and Parliament implicitly approves this division of responsibilities, it would be able to “abdicate” its law-making functions by delegating them to a person or entity who is responsible to no one.

Ultimately, the goal is to ensure that all functions of government—including the creation of delegated legislation—are subject to legislative control with the creation of a focal point for democratic accountability. Pushed too far, delegating broad legislative power to an independent agency upsets the balance in the system of parliamentary sovereignty and diverts democratic accountability away from ministers. This, I argue, is the mischief protected against by the non- abdication rule. Unlike a scope-based non-delegation rule, it directly prohibits the “eluding or

191 Ibid.

192 Sheehy & Feaver, supra note 19 at 184. See also Donald J Savoie, Court Government

and the Collapse of Accountability in Canada and the United Kingdom (Toronto:

University of Toronto Press, 2008) at 31.

deflecting [of] responsibility and thus electoral accountability.”193 This is arguably prescribed by the holdings and dicta in the non-abdication cases.

Before turning to questions of application, it is important to address a potential criticism concerning justiciability. How does a court determine whether a statute has the proper “control” mechanisms in place? This is already an issue in the context of the American non-delegation doctrine.194 In certain scenarios, statutory signals could point in multiple directions. Consider the

Telecommunications Act,195 which empowers the Canadian Radio-television and 2020 CanLIIDocs 693 Telecommunications Commission (“CRTC”). Under that statute, while the Cabinet may issue general policy directives to the CRTC196 and may, within time limits, “vary or rescind” a decision of the CRTC,197 the CRTC has broad powers to issue rules and orders that may have legislative effect without approval by ministers.198 At the same time, Cabinet can “issue to the

[CRTC] directions of general application on broad policy matters.”199 The responsible minister

193 Iancu, supra note 12 at 5.

194 Whittington & Iuliano, supra note 66 at 430: “Judges never developed the sort of

doctrinal tools that would allow them to meaningfully distinguish between inappropriate

abdication of legislative power and necessary delegation of administrative details.”

195 SC 1993, c 38.

196 Ibid, s 8.

197 Ibid, s 12(1).

198 Ibid, s 57.

199 Ibid, s 8.

can also issue a broad array of regulations on her own.200

Would a court find these mechanisms sufficient means by which to control the CRTC’s legislative power? The answer is not immediately obvious. The guiding principle, though, should be that a minister or Cabinet has the opportunity to control particular legislative acts completed by an “independent authority.” The question should simply be whether there is an operational possibility for a regulation to be made without an opportunity for ministerial control. In the

CRTC example, a regulation may be made without ministerial approval. The CRTC can make 2020 CanLIIDocs 693 regulations with legislative effect before a minister has a chance to approve them. This requirement of approval must find its way into any institutional instantiation of the non- abdication doctrine.

IV. APPLICATION AND A COMPLICATION

To illustrate how this doctrine would work in concrete cases and the institutional complications associated with it, the two cases introduced above are instructive: the Pan-Canadian Securities

Reference and the Saskatchewan Reference. These cases show the stakes of adopting a non- abdication rule on delegated power and how that rule could work. The non-abdication rule as described has no application in the Saskatchewan Reference because the delegation there was to the executive—even if the delegation was of a wide nature. However, the rule does have application in the context of the Pan-Canadian Securities Reference, where a new legislative body not contemplated by the Constitution Act, 1867 was envisioned by the legislative scheme.

200 Ibid, s 69.3(1).

A. SASKATCHEWAN REFERENCE

The Saskatchewan Reference involved the serious debate in Canada over the constitutionality of the federal carbon “tax.”201 One aspect of the case involved a lateral delegation from Parliament to the Cabinet of Canada. This delegation gave the Cabinet authority to amend the primary statute and adjust the rates payable by taxpayers as well as determine the conditions under which it applied.202 Specifically, the statute, as described by the Saskatchewan Court of Appeal, served the following purposes: 2020 CanLIIDocs 693 Part 1 implements a fuel charge. It applies to 22 GHG-producing fuels listed in

Schedule 2 of the Act. These include such things as gasoline, kerosene, fuel oil,

propane, coke oven gas, and methanol. The Governor in Council may add to, or

delete from, this list (s 3, definition of “fuel”; s 17(1)).203

This so-called “Henry VIII” power204 permits the executive “to amend by regulation the very statute which authorizes the regulation.”205 Such delegations are not clearly unconstitutional in

201 Saskatchewan Reference, supra note 20.

202 See ibid at para 101.

203 Ibid at para 36.

204 See Paul Daly, “Henry VIII Clauses in Comparative Perspective” (1 February 2017),

online (blog): Administrative Law Matters

comparative-perspective>, archived: .

205 See Ontario Public School Boards’ Assn. v Ontario (Attorney General), 151 DLR (4th)

346, 1997 CanLII 12352 (Ont SC) at para 50 [Public School Boards].

Canada and have, in fact, been upheld by courts.206

With this context in mind, the case raised two issues concerning delegation. First, there was a narrow issue of taxation and delegation, and whether the “fuel charge” in the statute was an unconstitutional “tax.” The second issue was the broader context of delegation and the propriety of using a Henry VIII clause to implement the statutory scheme. On the first issue of taxation, the power to tax ab initio in Canada cannot be assumed by the executive, as s. 53 of the

Constitution Act, 1867 states that all taxes must originate in the House of Commons.207 The 2020 CanLIIDocs 693 characterization of the “fuel charge” as either a tax or a regulatory charge was important because s. 53 would not apply to a regulatory charge.

In the Saskatchewan Reference, the majority of the Saskatchewan Court of Appeal ultimately concluded that the exaction was a regulatory charge, not a tax,208 and so it could be delegated. The majority, seemingly in response to arguments on the point, rejected the suggestion “that the delegation of authority to the Governor in Council in relation to the particulars of the Part 1 fuel charge [was] so wide that, while express, it nonetheless offend[ed] s.

53.”209 The majority concluded the power was merely ancillary in nature, such that it did not run afoul of the delegation restriction.210 Thus, in response to the potential claim that the delegation was “constitutionally problematic,” the majority of the Saskatchewan Court of Appeal was not

206 Ibid at para 54; Gray, supra note 13.

207 Eurig Estate (Re), [1998] 2 SCR 565, 1998 CanLII 801 at paras 31–32.

208 Saskatchewan Reference, supra note 20 at para 89.

209 Ibid at para 106.

210 Ibid at paras 102–104.

persuaded because the “scope” of the discretion was “confined.”211

The more interesting issue was the propriety of using a Henry VIII clause to enact the regulatory charge. The dissent went further on this ground, taking the issue out of the tax context and questioning whether the Henry VIII clause was constitutional at all because of the nature of the delegation and the principle of parliamentary sovereignty:

In fact, even if Part 1 were seen to be a regulatory levy, we would question the

constitutionality of the delegation of law-making power under it because s. 166(3) 2020 CanLIIDocs 693 sets forth a wholly-unfettered grant of broad discretion to amend Part 1 of the Act

that, by virtue of s. 166(4), may trump any exercise of legislative power by

Parliament. That is to say, ss. 166 to 168 expressly afford the executive branch of

government the power to circumvent the exercise of law-making power by the

legislative branch of government. This would seem unconstitutional, and contrary

to the Westminster model, regardless of where the so-delegated law-making

power lies under the Constitution Act, 1867.212

The majority’s entertainment of limits on delegation, and the dissent’s full acceptance of those limits, imply that there could be restrictions on the scope of the delegated power. The fault line between the majority and the dissent lies around the scope of the delegated power, particularly whether a broad power to amend laws is consistent with principles of parliamentary sovereignty and accountability.

How would the non-abdication rule apply to the Saskatchewan Reference? The non-

211 Ibid at para 105.

212 Ibid at para 372.

abdication rule and its limitations would apply to both issues raised in the case. On the majority’s analysis of the taxation issue, there was no need for it to specify that the power was “limited” or

“confined.” Under the non-abdication rule, there is no reason for courts to attach limits to the scope of delegation. Once the power was determined to be a regulatory charge, Parliament was free to delegate the power to levy those fees as it wished.

The more difficult issue in the case is the existence of the Henry VIII clause and the dissent’s analysis of that issue. The dissent framed the problem in terms of parliamentary 2020 CanLIIDocs 693 sovereignty.213 At first, this may appear to engage the non-abdication rule. After all, Henry VIII clauses “put the legislative onus in the executive rather than Parliament and for that reason, may be suspect.”214 However, while some courts have raised these concerns with Henry VIII clauses,215 the non-abdication rule would find them constitutionally unobjectionable in this context. While they permit the executive holding legislative power, Henry VIII clauses are merely a special case of the rule that Parliament may make “any law whatever,” and thus delegate any power whatever, subject to constitutional limits. The constitutional limits prescribed by the non-abdication rule attach to the recipient of the delegated power, but they do not discriminate as to the type of power that can be delegated. So long as there is a responsibility connection back to Parliament for the exercise of Henry VIII powers, there is no constitutional

213 Ibid at para 367ff.

214 Mark Mancini, “Abdicating Legislative Power: The Carbon Tax Case” (15 May 2019),

online: Advocates for the Rule of Law ,

archived: .

215 Public School Boards, supra note 205.

concern. In the carbon tax context, the connection exists: the executive is responsible for the powers exercised under Part I of the statute. In such cases, Parliament “could not abandon any of its own legislative jurisdiction”216 as it always has control over the Cabinet through the principle of responsible government. Henry VIII clauses, then, only speak to the type of power exercised by the executive. While these clauses may be undesirable from a good governance perspective,

Parliament always retains control over them.

To summarize, the Saskatchewan Reference is an interesting example that serves to make 2020 CanLIIDocs 693 the non-abdication rule concrete. Far from abdicating legislative power, Parliament transfers power to the executive subject always to the control embedded in the system of responsible government. While one might attack the scope or type of power delegated in this instance, that argument is not cognizable under the non-abdication rule. For that reason, the majority was correct in the Saskatchewan Reference.

B. PAN-CANADIAN SECURITIES REFERENCE

The tortured history of securities regulation in Canada presents a different kind of delegation norm with a different institutional mechanism. Because securities engage issues of provincial competence under the Constitution Act, 1867, efforts to regulate them nationally have historically encountered federalism problems, as interpreted by the Supreme Court of Canada in

Reference Re Securities Act.217 However, in Pan-Canadian Securities Reference, the federal government and select provincial and territorial governments attempted to get around the

216 Gray, supra note 13 at 170.

217 2011 SCC 66, [2011] 3 SCR 837.

federalism problems through a cooperative system with the following components:

1. a model provincial and territorial statute (“Model Provincial Act”) to be

mirrored by each participating province and territory;

2. a draft federal statute addressing systemic risk, criminal matters, and

national data collection;

3. a national regulator; and

4. the Council of Ministers, composed of the federal Minister of Finance and 2020 CanLIIDocs 693 the responsible ministers in each of the provinces, under whose

supervision the regulator would operate. 218

The Model Provincial Act and draft statutes in each case would have to be enacted into law by the respective legislatures: they would not have the force of law on their own.219 The Council of

Ministers could propose amendments to the Model Provincial Act, but the legislatures themselves would have to adopt the amendments.220 The regulator could adopt regulations that have the force of law, and any regulations proposed by the regulator would need to be approved by the Council of Ministers.221

218 Pan-Canadian Securities Reference, supra note 21 at para 21.

219 Ibid at para 24.

220 Ibid at para 25.

221 Ibid at paras 27–28; see also Canada, Minister of Finance, Capital Markets Stability Act

— Draft for Consultation (Ottawa: January 2016), s 76, online (pdf): Cooperative Capital

Markets Regulatory System

draft-revised-en.pdf>, archived: .

The scheme was ultimately challenged under the Constitution Act, 1867. The Attorneys

General of Quebec and Alberta argued that the scheme constituted a “transfer of legislative authority to the Council of Ministers which effectively create[d] a legislative body that is not contemplated by the Constitution.”222 The Supreme Court of Canada rejected this argument.223

To counter the claim that the scheme created a delegation of primary law-making power—the power to make, amend, or repeal statutes—the Supreme Court of Canada noted that all of the formative and substantive legislative acts on the part of the Council of Ministers are ultimately 2020 CanLIIDocs 693 subject to some form of legislative approval.224 As to the existence of the Council of Ministers, the Supreme Court of Canada noted that it was open to governments to create such a body, with the proviso that any amendments to the model/draft statutes are not implemented, as a matter of law, by the Council of Ministers, but instead by the responsible legislatures.225

The Supreme Court of Canada also held that the regulator was entitled to make regulations under the Council of Ministers’ supervision.226 The Court noted that “this form of delegation of administrative power is entirely consistent with the principle of parliamentary sovereignty, since the delegated authority can always be revoked by the sovereign legislature.”227

For the Supreme Court of Canada, “the legislature retains a significant degree of latitude in

222 Pan-Canadian Securities Reference, supra note 21 at para 40.

223 Ibid at para 80.

224 Ibid at paras 77–78.

225 Ibid at paras 78–79.

226 Ibid at para 121.

227 Ibid at para 123.

deciding how it will delegate administrative law-making powers pertaining to matters within its jurisdiction.”228 Short of federalism concerns, the Court repeated the statement that

“delegation…cannot impeach the assertion of power.”229

The argument advanced by the Attorneys General of Quebec and Alberta should be taken seriously under the non-abdication rule. The regulator and the Council of Ministers together create new law-making independent agencies, raising the prospect that a de facto new legislature has been created. This is not true with respect to the first part of the scheme challenged in the 2020 CanLIIDocs 693 case: the legislative acts that can be proposed by the Council of Ministers. Here, there is a responsibility link back to the relevant legislatures because substantive statutory amendments are subject to legislative approval.230 However, the making of regulations by the regulator appears to create a problem. Under the proposed federal statute, the Council of Ministers can approve or reject the regulations created by the regulator, but there is no mechanism for Parliament and the relevant legislatures to approve the regulations themselves.231 The provincial legislatures, as the

Attorney General of Quebec pointed out, also do not have input into the regulations adopted by the regulator.232 In other words, there is no control over some of the legislative actions taken by

228 Ibid at para 124.

229 Ibid at para 125, quoting Reference re Agricultural Products Marketing Association,

[1978] 2 SCR 1198 at 1225, 1978 CanLII 10.

230 See Pan-Canadian Securities Reference, supra note 21 at para 50.

231 See ibid at paras 27–28, 118–20.

232 See Pan-Canadian Securities Reference, supra note 21 (Factum of the Attorney General

of Quebec at paras 98ff).

the Council of Ministers. This is impermissible under the non-abdication rule.

Predictably, the Supreme Court of Canada met this concern by noting that “the delegated authority can always be revoked by the sovereign legislature.”233 This is cold comfort under the non-abdication rule. As noted above, in NS Interdelegation, Justice Rand alluded to the fact that the possibility of revocation is not sufficient to render a delegation constitutional.234 This applies just as well in the lateral delegation context as in the interdelegation context. The power to revoke is not a sufficient form of control if Parliament legitimately wishes to delegate power. 2020 CanLIIDocs 693 In this sense, the Pan-Canadian Securities Reference was incorrectly decided if the non- abdication rule is applied to the facts. Under the rule, there should be some mechanism through which the regulations produced by the regulator are ultimately subject to Parliamentary approval.

Under the cooperative federalism scheme, this does not appear to be the case. For that reason, the ability of the regulator to issue regulations should not be subject to the approval of the Council of

Ministers but should be subject to the approval of Parliament and the legislatures themselves.

C. INSTITUTIONAL COMPLICATIONS

Future scholarly efforts in this field may consider the institutional features of the problem. The non-abdication rule only provides that Parliament retain control over every instance of delegated law-making regardless of scope. However, the doctrine does not speak to how it can be institutionally instantiated, or in other words, how Parliament could control delegated law- making.

233 Ibid at para 123.

234 NS Interdelegation, supra note 13 at 50.

One way that Parliament already does so, as noted above, is through its control of ministers. This is the method described in the Pan-Canadian Securities Reference, particularly in reference to proposed statutory amendments. Parliament, through ministers, must approve them before they go into force. Hypothetically, however, Parliament could simply approve regulations issued by the regulator directly.

Another way that Parliament could control administrative decision makers is through general review, such as that adopted under the Statutory Instruments Act.235 This legislation 2020 CanLIIDocs 693 allows a committee of Parliament to scrutinize regulations on defined grounds.236 Parliament exercises supervision over independent agencies directly instead of through ministers. Because

Parliament has delegated around ministers, ministers are cut out of the equation. This parliamentary innovation could be seen as an attempt to meet the non-abdication rule. Indeed, as

Minister of Justice John Turner said when the SIA was adopted, the increase of volume of delegated legislation warranted a political solution in order to exert parliamentary power.237

The relative merits of a direct ministerial method or the parliamentary method are beyond the scope of this paper. But to give a flavour of some of the complications, the

SIA, at least to some extent, has not been the model of monitoring. The Standing Joint

Committee for the Scrutiny of Regulations (the “Committee”), a Parliamentary committee, operates under the SIA and has wide scope to review all statutory

235 RSC 1985, c S-22 [SIA].

236 See ibid, s 19.

237 “Bill C-182, Statutory Instruments Act”, 2nd reading, House of Commons Debates, 28-3,

vol 3 (25 January 1971) at 2734–37.

instruments.238 The term “statutory instruments” is broadly defined.239 Under the legislation, only regulations (as opposed to rules, guidelines, or decisions of a legislative nature) are published in the Canada Gazette, yet all statutory instruments stand permanently referred to the

Committee.240 It is beyond the scope of this paper to discuss the various nuances of the statutory scheme. But for some critics, the SIA and the mechanisms it set in place241 have failed to adequately control the delegation of power,242 let alone meet the standards of the non-abdication rule. 2020 CanLIIDocs 693 The problems with the SIA are many. One is with the legislation itself. What constitutes a

“statutory instrument” that must be approved by Parliament under the statute is honey-combed with exceptions.243 Furthermore, under the SIA, what constitutes a “statutory instrument” is apparently dependent on how the Department of Justice interprets that term.244 If the legislature,

238 See SIA, supra note 235, s 19.

239 See ibid, s 2(1).

240 See ibid, s 19.

241 I ignore here the various mechanisms in place for the “publication and scrutiny” of

regulations. The SIA does impose a number of requirements for public scrutiny of

delegated legislation like regulations. But the requirements of publication are somewhat

different to what I am concerned with here: Parliamentary control over delegated

legislation.

242 Neudorf, supra note 1 at 562.

243 See SIA, supra note 235, s 2(1).

244 Denys C Holland & John P McGowan, Delegated Legislation in Canada (Toronto:

in delegating power, does not include a certain set of words, then a particular legislative act will not be subject to any of the terms under the SIA, or potentially any review whatsoever, unless an enabling statute permits the minister to conduct such a review.245 Then, even if the Committee does get to review a particular legislative act, it only does so on certain technical grounds.246 It does not review on policy grounds, for example, whether the rule is a good idea at all.247 All of this limits the potential legislative instruments that are scrutinized by Parliament.

Nonetheless, one can see how a non-abdication rule could work, despite the institutional 2020 CanLIIDocs 693 complications. When Parliament delegates to a minister or cabinet directly, as in the

Saskatchewan Reference, no abdication problem arises because responsible government continues to exist, regardless of whether it is efficacious. On the other hand, when Parliament delegates to an independent agency like the Council of Ministers in the Pan-Canadian Securities

Reference, one must take a harder look at the statute enabling the agency and other statutes bearing on the problem, including the SIA. If a court considers that every legislative act adopted by that agency is somehow traceable back to Parliament, either through ministers or to

Parliament directly, then no abdication problem arises. The problem for a court will be to determine whether statutes like the SIA, which at first do not appear sufficient to control delegated power (because not every legislative act is subject to its auspices), can nonetheless be sufficient in particular circumstances. Nevertheless, the basic question is whether Parliament has

Carswell, 1989) at 12–13.

245 Ibid at 13.

246 Neudorf, supra note 1 at 564.

247 Ibid.

approval over every instance of delegated law-making.

V. CONCLUSION

I have argued that while a “non-delegation” doctrine based on the scope of delegation is ill-fitted to the Canadian-Westminster system of government, an ex post rule of “non-abdication” has been recognized by the Supreme Court of Canada. That doctrine is arguably supported by the principles of parliamentary sovereignty, responsible government, and a more general norm of 2020 CanLIIDocs 693 political accountability in which the law-making functions of the state are subject to some control by Parliament and thus the people. The central idea is that Parliament cannot create new law-making bodies under the Constitution without establishing control over that delegated power.

At all times, legislative delegation must be squared with the Constitution. While, as cases referenced throughout this article note, it is clearly open to Parliament to seek the assistance of administrative actors, it is not open to Parliament to divert accountability for law-making in the guise of “independence.” This is a tendency that should be guarded against in the constant struggle to ensure the consistency of the administrative state with fundamental constitutional norms.