November 10, 2003

Mr. Michael Phelps
Chair
WPC – Committee to Review the Structure of
Securities Regulation in Canada
25th Floor
700 West Georgia Street
Vancouver, British Columbia
V7Y 1B3

Dear Mr. Phelps:

Re:

Constitutional authority to implement the CSC model

 

You have asked my opinion on two constitutional questions relating to the Canadian Securities Commission (“CSC”) model outlined in Chapter 6 of the Committee’s report.  The two questions, and a summary of my opinion concerning them, are as follows.

1.       Does the federal government have the legislative authority to enact legislation implementing the CSC model?  Do the provinces and territories correspondingly have legislative authority to enact legislation that: (a) incorporates by reference the new federal securities legislation; (b) delegates administrative powers to the CSC; and (c) dissolves their respective existing provincial and territorial securities regulators?

The federal government has legislative authority under its general regulation of trade power, conferred by s. 91(2) of the Constitution Act, 1867, to implement the CSC model.  Based on the factors that the Supreme Court has identified as relevant, the CSC model would constitute a valid general regulation of trade rather than an impermissible regulation of a particular trade.  The provinces and territories may enact the corresponding legislation to which this question refers.  Legislatures are afforded wide scope to enact incorporations by reference and administrative interdelegations.  The power to establish a regulatory body includes the power to dissolve it.

2.       If one or more provinces or territories decide not to enact the legislation described above, and the federal government concludes that this would jeopardize the successful operation of the scheme in other parts of the country, would the federal government nevertheless have the legislative authority to implement the CSC model on a national basis by including in its legislation an express paramountcy clause that plainly demonstrates an intention to enact a complete code and which states that federal legislation alone would govern, to the exclusion of provincial and territorial regulation, on prescribed matters, including amongst other matters: (a) prospectus review; (b) registration of market intermediaries; (c) take-over bid regulation; (d) continuous disclosure by public companies: (e) prospectus exempt financings (f) international securities agreements; and (g) securities regulatory enforcement actions?

Federal legislation that included an express paramountcy clause of this kind would be effective to exclude the operation of provincial and territorial regulation of the matters to which the paramountcy clause applied.  As currently interpreted, the paramountcy doctrine renders provincial legislation inoperative where the application of the provincial legislation would displace the legislative purpose of Parliament.  While it is not necessary that it do so, Parliament is entitled to state expressly its purpose that the CSC model, and it alone, should govern specified aspects of securities regulation.

I will now set out the basis for my opinion, dealing with the two questions in turn.

1          Federal authority to implement, and provincial and territorial authority to defer to, the CSC model

Federal authority to implement the CSC model

In my view, the federal government has authority under its general regulation of trade power, conferred by s. 91(2) of the Constitution Act, 1867, to implement the CSC model.  It is also arguable that the federal power to legislate on matters of “national concern”, recognized as stemming from the “peace, order and good government” clause in the opening words of s. 91 of the Constitution Act, 1867, would support implementation of the model.  Other federal powers would be available to support legislation implementing portions of the model.  But since the federal government could in my view rely on the general regulation of trade power in implementing the model in its entirety, I will focus my discussion on that power.

The federal general regulation of trade power

Section 91(2) of the Constitution Act, 1867 gives the federal government a power described as “The Regulation of Trade and Commerce”.  Since the decision in Citizens Insurance Co. v. Parsons (1881), 7 App. Cas. 96, it has been settled constitutional law that the power conferred by s. 91(2) has two branches:  (1) the regulation of interprovincial and international trade, and (2) “the general regulation of trade affecting the whole dominion”.  It has also been settled that s. 91(2) does not authorize regulation of the contracts of a particular business or trade.  That power rests with the provinces under s. 92(13) as a matter of property and civil rights.

The general regulation of trade power lay largely unused, and its content largely unexplored, for many years following the Parsons case.  However, in more recent cases the Supreme Court of Canada has clarified the scope of the power.  Where it applies, it authorizes federal regulation not only of interprovincial and international trade, but also of trade within a province.

The leading case on the scope of the power is General Motors of Canada Ltd. v. City National Leasing Ltd., [1989] 1 S.C.R. 641.  According to the Court’s reasons in that case, five factors are relevant in determining whether federal legislation qualifies as a general regulation of trade.  These factors also assist in making this determination in a way that maintains an appropriate balance between federal power under s. 91(2) and the provinces’ property and civil rights power under s. 92(13).  The five factors are:

(1)              the legislation must be part of a general regulatory scheme;

(2)              the scheme must be monitored by a regulatory agency;

(3)              the legislation must be concerned with trade as a whole rather than with a particular industry;

(4)              the legislation should be of a nature that the provinces jointly or severally would be constitutionally incapable of enacting; and

(5)              the failure to include one or more provinces or localities in a legislative scheme would jeopardize the successful operation of the scheme in other parts of the country.

The Court stated that these five factors are not exhaustive, and that the presence or absence of any of them will not necessarily be determinative.  However, their presence will “make it far more probable that what is being addressed in a federal enactment is genuinely a national economic concern and not just a collection of local ones”.

In the General Motors case, the Supreme Court applied these factors to uphold the federal Combines Investigation Act (now the Competition Act) as a general regulation of trade.  It had no difficulty finding a regulatory scheme and oversight by a regulatory agency, the Director of Investigation and Research (now the Commissioner of Competition), so that the first two factors were present.  It described the last three factors as sharing “a common theme:  all three are indications that the scheme of regulation is national in scope and that local regulation would be inadequate”.  In concluding that the third factor was present, it focused on the generality of application of the Combines Investigation Act.  It described the Act as “quite clearly concerned with the regulation of trade in general, rather than with the regulation of a particular industry or commodity”, and as “aimed at eliminating commercial practices which are contrary to healthy competition across the country, and not in a specific place, in a specific business or industry”.

In addressing the last two factors, the Court reviewed what it described as “the diverse economic, geographical, and political factors which make it essential that competition be regulated on the federal level”.  These included the increased mobility of labour, capital, goods and technology across provincial boundaries, facilitated by improvements in communications and transportation, and the need for national competition policy to promote and protect the national common market and the welfare of the national economy, free from the distortions that might result from differing provincial competition policies.  It saw it as “evident ... that competition cannot be effectively regulated unless it is regulated nationally”. 

For the same reasons, the Court rejected the argument that the legislation should be read down so as not to apply to intraprovincial trade.  Intraprovincial trade could not be excluded both “[b]ecause regulation of competition is so clearly of national interest and because competition cannot be successfully regulated by federal legislation which is restricted to interprovincial trade”.  It is clear, therefore, that where federal legislation qualifies as a general regulation of trade, it can validly extend not only to interprovincial and international but also to intraprovincial trade.

The CSC model as a general regulation of trade

Based on the factors discussed by the Supreme Court in General Motors and the approach to the federal general regulation of trade power that they express, I conclude that federal legislation implementing the CSC model would be constitutionally valid as a general regulation of trade. 

The first two factors to which the Supreme Court referred in General Motors – that the legislation be part of a general regulatory scheme and that the scheme be monitored by a regulatory agency – would readily be met.  The remaining three factors require more detailed discussion.

In my view, legislation implementing the CSC model would properly be described as concerned with trade as a whole rather than with a particular trade or industry, so as to meet the third factor.  While it could be argued that the regulation was of a particular trade or industry – the securities trade or industry – that characterization would not accurately capture what the model would entail. 

It is true that provincial securities legislation has often been characterized as regulating the securities trade.  This characterization fits well with the constitutional basis, s. 92(13) of the Constitution Act, 1867, invoked to support provincial legislation:  s. 92(13) has long been held to confer on the provinces the power to regulate particular trades or industries.  But the CSC model would extend to a great many participants in the capital markets other than dealers, brokers and others who can be regarded as involved in the securities trade or industry per se.  In extending to issuers, for example, it would apply to market participants engaged in a wide variety of businesses and trades – to all those, whatever their ordinary business or trade, who issue and seek to raise capital in Canada on the strength of securities.  In this respect the model can properly be analogized to the generally applicable scheme regulating competition upheld in General Motors.

The fourth factor, again, is that the provinces jointly or severally would be constitutionally incapable of enacting the legislation.  The Supreme Court did not fully spell out the meaning of the fourth factor in General Motors.  It focused more on the fifth factor– whether the failure to include one or more provinces or localities would jeopardize the successful operation of the scheme in other parts of the country – and on the question whether provincial regulation of competition could be effective, than on the question of constitutional incapability.  However, its acknowledgement that provinces can and do pass their own competition laws makes it clear that the absence of any provincial power at all in the field is not required for the fourth factor to be present.  Although it was not explicit in General Motors, the key to the Court’s conclusion on the fourth factor seemed to be that the increasingly interprovincial nature of the flow of goods, services and capital precluded the provinces from effectively setting competition policy and regulating competition.

A variety of means have been suggested through which the provinces can enhance the cooperation that already exists between them in securities regulation and deploy their powers to create a virtual scheme of national regulation.  But this does not necessarily mean that the constitutional incapability factor is absent.  The Supreme Court’s recent decision in Unifund Assurance Co. v. Insurance Corp. of British Columbia, 2003 SCC 40, emphasizes that a province may only exercise its regulatory authority over people and matters with a sufficient connection to the province.  The Committee’s research, as I understand it, strongly suggests that capital markets have become increasingly integrated, and that Canada’s capital markets have become national in scope.  It can readily be argued that separate provincial regulatory regimes, each subject to territorial limits, are constitutionally incapable of comprehensively regulating securities, just as they could not effectively regulate competition.

The fifth factor – the potential impact of failure to include one or more provinces or localities on other parts of the country – encompasses in the securities context, in my view, the potential consequences of regulatory failure in one province for capital markets and participants in capital markets elsewhere in Canada.  There appears to be a sound basis for concluding that this factor is present.  For example, one province’s failure to adopt an adequate regulatory scheme could both impair the reputation of Canadian capital markets generally, and adversely affect the residents of other provinces whose securities are traded there.

According to the decision in General Motors, the presence of all of the factors is not strictly necessary for legislation to constitute a general regulation of trade.  The underlying purpose of all of the factors is to help determine whether what the scheme is addressing is genuinely a national economic concern and not just a collection of local ones.  The last three factors go to whether the scheme of regulation is national in scope and whether local regulation would be inadequate.

Upholding the CSC model as a general regulation of trade could therefore ultimately turn on how the courts approached the two overarching questions of national scope and adequacy or effectiveness of provincial regulation.  The answer to the national scope question would likely favour upholding the model.  The Supreme Court has shown itself prepared, in cases like Multiple Access Ltd. v. McCutcheon, [1982] 2 S.C.R. 161, and Global Securities Corp. v. British Columbia (Securities Commission), [2000] 1 S.C.R. 494, to recognize that there is a strong transprovincial and transnational dimension to Canadian capital markets.  The Committee’s conclusions appear to provide further support for this assessment.  The CSC model would be general and national, not particular and local, in its application.

The Court’s repeated references in General Motors to the need for federal regulation if there is to be effective regulation indicate that the question of effective regulation might well be determinative.  The way in which the general regulation of trade power has developed thus places the Court in a somewhat unusual position.  Ordinarily, as the Supreme Court stated in Reference re Firearms Act (Can.), [2000] 1 S.C.R. 783, the Court does not evaluate the efficacy – or lack of efficacy – of legislation in deciding constitutional division of powers questions.  Ordinarily, for a number of institutional reasons, it is not for judges but for legislators to gauge the effectiveness of legislation.

In the General Motors case, the Supreme Court paid heed in considering the effectiveness question to the conclusions of studies that formed part of the background to the enactment of federal competition legislation.  It was appropriate in my view for the Court to do so.  It would be equally appropriate in considering the effectiveness question in relation to federal legislation implementing the CSC model for courts to take account of your Committee’s conclusions and recommendations, along with other views considered by Parliament before enacting the legislation.  There will almost certainly continue to be a range of opinions on the effectiveness question, as there has been to date.  But particularly if the material put before it gave Parliament a reasonable basis to conclude that federal securities regulation is necessary to ensure effective regulation, it is likely in my view that legislation implementing the CSC model would be held constitutionally valid as a general regulation of trade.

Provincial and territorial authority to defer to the CSC model

Once the federal legislation implementing the CSC model was enacted, in my view the provinces would be fully entitled to enact legislation incorporating the new federal securities legislation by reference, delegating administrative powers to the CSC and dissolving their respective securities regulators. 

Because the territorial legislatures do not have independent constitutional status, and the exercise of their legislative authority is subject to federal legislation, the federal government would have authority to decide how best to deal with securities regulation in the territories following the implementation of the CSC model.  I will therefore confine the balance of this discussion to the provinces’ authority to defer to the CSC model.

The enactment of the federal legislation would not by itself preclude the provinces from continuing to regulate securities.  They could continue to do so.  The federal power to enact legislation that constitutes a general regulation of trade does not take away the provinces’ regulatory powers.  Whether their legislation would continue to operate if they chose not to defer to the new federal legislation would depend on the application of the paramountcy doctrine, discussed below in answer to the Committee’s second question.

But there would be no constitutional difficulty with their exercising their powers, if they chose to do so, by referentially incorporating the federal legislation and delegating administrative powers to the CSC.  It is constitutionally impermissible for the federal Parliament and provincial legislatures to delegate legislative authority directly to each other.  But as the Supreme Court of Canada confirmed in R. v. Furtney, [1991] 3 S.C.R. 89, “Parliament may delegate legislative authority to bodies other than provincial legislatures, it may incorporate provincial legislation by reference and it may limit the reach of its legislation by a condition, namely the existence of provincial legislation.”  Provincial legislatures have corresponding authority to delegate to other bodies, to incorporate federal legislation by reference and to limit the reach of provincial legislation by conditioning its application on federal legislation.

Incorporation by reference entails enacting legislation providing that a matter within the legislature’s authority be governed by legislation enacted by another legislature, either as it exists at the time of the incorporation or, where the incorporating legislation is explicit on this point, as it might be modified in the future.  This is regarded as one legislature’s adopting as its own and for its own purposes legislation enacted by another.  It is very common in Canada to combine this technique with the delegation of regulatory authority to a regulatory body created by another legislature, coupled with a direction to that regulatory body to apply that legislature’s law. 

While there are suggestions in the case law that the power of delegation is subject to “reasonable limits”, and that Parliament or a legislature may not go so far as to “abdicate” or “abandon” its powers to a delegate, very broad delegations of authority have repeatedly been upheld.  These include delegations that give regulatory bodies extensive powers to fashion regulatory schemes governing particular areas of economic activity.  The courts have appreciated the desirability of intergovernmental cooperation in enacting comprehensive schemes of regulation.  In my view it is very unlikely that any delegation of authority to regulate securities would be regarded as an impermissible abdication or abandonment of legislative power.

There would also, in my view, be no constitutional difficulty with dissolving existing regulators.  As a matter of legislative sovereignty, the power to enact legislation includes the power to repeal it.  Legislation that created a regulatory body could be amended or repealed to dissolve it.

2.         Federal authority to exclude provincial and territorial regulation through paramountcy

In my view the federal government would be entitled to ensure that the CSC model operated as a complete code governing specified securities matters by including in the implementing legislation an express paramountcy clause. Federal legislation would be effective to exclude provincial securities regulation under the paramountcy doctrine if it plainly demonstrated an intention that it and it alone govern the matters from which provincial regulation was to be excluded. An express paramountcy clause would constitute an unmistakable, and in my view constitutionally effective, expression of this intention.

Because of the territories’ dependant constitutional status, there would be no need to invoke the paramountcy doctrine to exclude the operation of territorial securities regulation.  The federal Parliament could directly override territorial legislation.

The paramountcy doctrine applies where there is conflict in operation between valid provincial and valid federal legislation.  Its result is that the federal legislation prevails, and the provincial legislation is inoperative, to the extent of the conflict.

The key determinant of whether the paramountcy doctrine applies is the test for identifying a conflict in operation between the federal and the provincial legislation.  As the Supreme Court’s recent decisions make clear, the test now applied is whether operation of the provincial legislation “would displace the legislative purpose of Parliament”.

That phrase first appeared in Multiple Access Ltd. v. McCutcheon, where the Supreme Court appeared to adopt a stringent test of express contradiction for determining the existence of a conflict sufficient to trigger operation of the paramountcy doctrine.  Paramountcy was invoked in that case to argue that federal insider trading legislation, applicable to federally incorporated companies, rendered inoperative the virtually identical insider trading provisions of the Ontario Securities Act

The Supreme Court held that the paramountcy doctrine did not apply.  It stated the test as one of “actual conflict in operation as where one enactment says ‘yes’ and the other says ‘no’; ‘the same citizens are being told to do inconsistent things’; compliance with one is defiance of the other”.  Duplicate legislation did not present a conflict based on this test.  Rather, duplication was “the ultimate in harmony”.  The Court added that because the two sets of legislation were so similar, “the legislative purpose of Parliament will be fulfilled regardless of which statute is invoked by a remedy-seeker; application of the provincial law does not displace the legislative purpose of Parliament.”  The Court’s inference was that Parliament did not intend that only its legislation govern.

Three more recent Supreme Court decisions signal a less restrictive approach to identifying a conflict, one that builds on the reference in Multiple Access to “displacing the legislative purpose of Parliament”, and gives primacy to Parliament’s intention as to whether its legislation should operate as a complete code.

In Bank of Montreal v. Hall, [1990] 1 S.C.R. 121, the Court held that provincial legislation setting conditions on secured creditors’ enforcement of security interests conflicted with and was rendered inoperative by federal legislation governing Bank Act security.  The federal statute gave a bank holding the security an immediate right to realize on it on default.  The Court rejected the argument that there was no conflict in operation because banks could simply follow the provincial legislation.  “The focus of the inquiry,” it stated, “must be on the broader question whether operation of the provincial Act is compatible with the federal legislative purpose.”  “Dual compliance will be impossible when application of the provincial statute can fairly be said to frustrate Parliament’s legislative purpose.”  The Court summarized the situation as one in which Parliament had “enacted a complete code”, so that there was “no room left for the operation of the provincial legislation”.

In M & D Farm Ltd. v. Manitoba Agricultural Credit Corp., [1999] 2 S.C.R. 961, the Court referred to the “extended meaning” given by the relevant jurisprudence to the “express contradiction” between federal and provincial statutes necessary to trigger the paramountcy doctrine.  In that case a farmer had obtained a stay of proceedings by creditors under federal legislation.  As permitted by provincial legislation, a mortgagee then obtained leave from the court to commence foreclosure proceedings.  Though the creditor could comply with both statutes by not taking enforcement measures, the Court found an operational conflict:  the legal system could not simultaneously create an entitlement to enforce and a prohibition on enforcement.  The Court was also prepared to look to the substance rather than the form of the two enactments – one of which directly stayed creditor enforcement actions and the other of which required leave to commence enforcement proceedings – to recast them as creating conflicting directives to the court dealing with enforcement of the mortgage.

In Law Society of British Columbia v. Mangat, [2001] 3 S.C.R. 113, the Court again emphasized the importance of looking, in determining whether a conflict in operation exists, at “whether the application of the provincial law will displace the legislative purpose of Parliament”.  The Court found an operational conflict where a provincial law prohibited non-lawyers from acting in immigration proceedings, while the federal law authorized non-lawyers to appear and charge a fee.

The Court dismissed as “superficial” the suggestion that there was no operational conflict because a person could comply with both enactments, either by becoming a lawyer or by not charging a fee.  While complying with the stricter provincial statute would necessarily involve complying with the federal statute, dual compliance was impossible based on the “expanded interpretation” of conflict in M & D Farm and Bank of Montreal v. Hall.  “Where there is an enabling federal law,” the Court stated, “the provincial law cannot be contrary to Parliament’s purpose.”

In my view, the current focus in the paramountcy cases on the purpose of Parliament means that the federal government could, through legislation implementing the CSC model, preclude the operation of provincial securities regulation.  It would not be strictly necessary in my view to include an express paramountcy clause to bring about this result.  It would be sufficient that the legislation plainly manifested an intention that it and it alone govern the specified aspects of the transactions and activities to which it applied.  If this intention was apparent, then a court should conclude, whatever the degree of similarity or duplication between the federal and the provincial legislation, that the operation of the provincial legislation would “displace the legislative purpose of Parliament”, so that the paramountcy doctrine would render it inoperative.

While an express paramountcy clause would not be necessary if this intention was otherwise clearly expressed, it would in my view be open to Parliament to include an express paramountcy clause specifying that the federal legislation alone should govern.  An express paramountcy clause would, among other things, make it clear that Parliament’s intention was not the same as that expressed in the legislation considered in the Multiple Access case.  The question of paramountcy of the CSC model would arise in a very different legislative and policy context from that of the Multiple Access case.  But because that case was a securities case, in which the Supreme Court applied the paramountcy doctrine so as to permit federal and provincial regulatory schemes to operate concurrently, a court might in the absence of an express paramountcy clause be tempted to follow Multiple Access and hold that both sets of regulations could operate.  An express paramountcy clause would constitute an unmistakable expression that the continued operation of provincial regulation in the face of the CSC model would “displace the legislative intention of Parliament”.

Including express paramountcy clauses has not been the ordinary practice in federal legislative drafting.  As a result, courts have typically been left to infer Parliament’s legislative purpose from the operative terms of the legislation.  But I see no reason in principle why Parliament should not be entitled to make its purpose express, and why the courts should not give effect to an express paramountcy clause in federal securities legislation. 

As I understand it, the Committee’s second question contemplates an express paramountcy clause that would not apply generally to the legislation implementing the CSC model, and would thus not seek to exclude all aspects of provincial securities regulation.  Instead, it would cover specified aspects of securities regulation, from which Parliament considered it necessary to exclude provincial regulation to avoid jeopardizing the operation of the federal scheme. 

Framing an express paramountcy clause in this way would in my view enhance the likelihood that the courts would accept this technique.  An approach along these lines should not as a legal matter be required for an express paramountcy clause to be given effect.  However, a more focused rather than a more general clause might as a practical matter be regarded as more palatable, because it would allay possible concerns about the extent of the impact of paramountcy on a recognized field of provincial regulation.

Yours very truly,

John B. Laskin

JBL/tp