At issue in Boulter was the constitutional validity of legislation requiring the Utility and Review Board to set the same power rates for all consumers. The appellants argued poverty was an analogous ground under s. 15(1) of the Canadian Charter of Rights and Freedoms, and the legislation discriminated against the poor. In dismissing the appeal, the Court of Appeal held that poverty was neither an immutable characteristic of the individual nor constructively immutable, and thus did not meet the criteria to be considered an analogous ground.

While an argument can be made that poverty can be seen as constructively immutable, (see Daniel Del Gobbo, (Mis)Construing Kapp to Preclude the Impoverished,) the entire immutable characteristics doctrine is somewhat puzzling. The purported policy rationale of s. 15 is to promote equality by preventing legislation being passed that exacerbates existing inequalities. In Law v. Canada, [1999] 1 S.C.R. 497, the Supreme Court of Canada set out the tests for determining a discrimination claim. Under the first branch of the Law test, one must inquire as to whether the impugned law draws a formal distinction between the claimant and others on the basis of one or more personal characteristics. In Corbiere v. Canada, [1999] 2 S.C.R. 203, the Court established the criteria to identify an analogous ground. The Court held that the common thread running through all enumerated grounds in s. 15 is that they serve as the basis for stereotypical decisions made on the basis of a personal characteristic that is immutable, or changeable only at great cost to personal identity.

The decision to limit the scope of s. 15 to discrimination based on personal characteristics has a valid objective. As McIntyre J. wrote in Andrews v. Law Society of British Columbia, [1989] 1 S.C.R. 143:

Distinctions based upon personal characteristics attributed to an individual solely on the basis of association with a group will rarely escape the charge of discrimination, while those based on an individual’s merits and capacities will rarely be so classed.

The purpose of such enumerated and analogous grounds, then, is to serve as an indicator for suspect decision making. The adoption of “personal characteristics” serves as a bright line, clearly demarcating between decisions based on stereotypical decisions and those based upon merit. However, as Professor Alan Dale Gibson points out in “Analogous Grounds of Discrimination Under the Canadian Charter: Too Much Ado About Next to Nothing”, (1991) 29 Alta. L. Rev. 772, the “personal characteristic” standard is inherently incompatible with the notion of discrimination. Discrimination is based upon broad stereotypes of a group, rather than one’s personal characteristics. The Court recognized such difficulties in Corbiere when it introduced the concept of a “constructively immutable” characteristic; thus, grounds such as marriage and religion which would not be easily encompassed by the umbrella of immutable “personal characteristics” can nonetheless fit under the Court’s expanding definition of what may constitute discrimination. Yet the Court relegates other factors such as personal care and social assistance to the side, despite their potential to serve as bases for discriminatory decision making.

In adopting the “personal characteristics” standard as an indicator for discrimination, the Court runs the risk of ignoring discrimination on other grounds. The real danger lies in relying on preconceived notions of what may constitute discrimination, and dismissing actual discrimination where it doesn’t fit tidily in our prior definitions. One is well served to remember Wilson J.’s caution in R. v. Turpin, [1989] 1 S.C.R. 1296:

If the larger context is not examined, the s. 15(1) analysis may become a mechanical and sterile categorization process conducted entirely within the four corners of the impugned legislation.

Returned to Boulter, one is struck by the Court’s repeated refusal to recognize poverty as an analogous ground of discrimination. In Sparks v. Dartmouth/Halifax County Regional Housing Authority, 1993 CanLII 3176 (NS C.A.), the court largely avoided the poverty argument in favour of finding the ground of distinction was “public housing tenant”. While the end result may have been the same, the Court’s refusal to extend its analysis beyond the confines of the narrow boundaries it itself has created is troubling. What would be the difference if an individual is denied tenancy on the basis of their race or ethnicity instead of on the basis of their impoverished state? In both situations the individual would be experiencing discrimination on the basis of a stereotype about a particular group. Are we prepared only to recognize discrimination when it occurs to members of a sub-group on a recognized analogous ground, denying s. 15 protection to other segments of the population?

Whether or not poverty should be considered an analogous ground, and whether the legislation in Boulter is indeed discriminatory is not at issue here. The refusal to consider poverty as an analogous ground since it fails to meet the “personal characteristics” standard, however, is a matter of some concern. The “personal characteristics” standard has the effect of drastically reducing the scope of s. 15 protection. Though useful as an indicator of discrimination, its use as a gatekeeper function for s. 15 defeats its very purpose.

In reconceptualising the criteria to establish analogous grounds, perhaps the “personal characteristics” standard could be replaced with a more flexible approach that has the potential to encompass all grounds of discrimination, regardless of whether they conform to our preconceived norms. For example, the “distinct community” standard in Commission For Racial Equality v. Dutton, [1989] QB 783, could theoretically include poverty as an analogous ground. Perhaps we can also rephrase McIntyre J.’s explanation of discrimination in light of our expanding conceptions of what constitutes discrimination.

Distinctions based upon stereotypes of an individual on the basis of association with a group will rarely escape the charge of discrimination, while those based on an individual’s merits and capacities will rarely be so classed.

While this would expand the scope of s. 15 protection, this is by no means a bad thing. The focus on stereotypes stemming from one’s association with a group acts both to encompass other discriminations and specify a specific and identifiable disadvantaged group. It is indeed the collective characteristics of the group –be it gender, sexual orientation, or religion– that is the target of the stereotypical assumptions. In removing the “personal characteristics” standard, it is our hope that an analysis for analogous grounds can take place free from our own myths and stereotypes of what may constitute discrimination.

Background

In 1940, the Constitution Act, 1867 was amended to give Parliament jurisdiction over unemployment insurance (later referred to as employment insurance), as a somewhat-belated response to the economic and social effects of the Great Depression. Since then, the federal government has maintained an employment insurance system paid for by premiums levied on employment income.

After 1980, however, the system began to run a deficit. In 1996, Parliament reformed the system by passing the Employment Insurance Act, which allowed the government to use money collected under EI to fund “active measures” that aimed to reduce unemployment and help affected persons re-enter the workforce. Specifically, the Act created employment benefits programs that helped create jobs and encouraged unemployed persons to accept work.

The Act also allowed the Employment Insurance Commission to set premium rates at a level that ensured there would be enough money each year to pay EI benefits and that raised a reserve in the event of an economic downturn. This premium-setting mechanism was amended in 2001 and 2004 to allow the Cabinet to adjust the premium directly.

The reserve created by the Employment Insurance Act has grown considerably in the past 12 years and is now in excess of $50 billion. The size of this reserve, and allegations of its misuse, convinced the appellant labour organizations to ask the Court to declare several aspects of the employment insurance program unconstitutional.

Issues Before The Court

The appellants challenged the constitutional validity of two aspects of the Employment Insurance Act. First, they argued that Parliament’s jurisdiction over unemployment insurance, pursuant to section 91(2A) of the Constitution Act, 1867, is limited to compensating unemployed persons, and that any use of EI funds to implement measures beyond compensation is ultra vires Parliament’s s. 91(2A) power. The second issue concerned the Act‘s premium-setting mechanism. The appellants claimed that s. 91(2A) of the Constitution only allows Parliament to collect enough funds to maintain the EI program and to build a reasonable reserve, but that the Act’s impugned method of setting annual premium rates allows the government to collect beyond what is necessary.

Employment Insurance Includes Unemployment Prevention

Writing for a unanimous Court, Justice Lebel rejected the appellants’ claim that Parliament’s unemployment insurance power, as provided by section 91(2A) of the Constitution, is limited to compensating wages during periods of unemployment. In his reasons, Justice Lebel makes it clear that Parliament’s s. 91(2A) power must be given a generous and purposive interpretation. He states that the objectives behind the s. 91(2A) power are to alleviate the negative effects of unemployment and to ensure that unemployed persons maintain ties to the labour market such that they can eventually re-enter the workforce. Any federal initiative that is directed towards one of these two objectives is a valid use of the section 91(2A) power. He writes:

Regulating unemployment insurance does not mean simply taking passive responsibility for paying benefits to Canadian workers during periods when they are not working. It also means taking on a more active role designed to maintain or restore ties between persons who may become or are unemployed and the labour market. [official English translation]

The employment benefits programs created by the Employment Insurance Act – which include such measures as supplementing a worker’s income if they accept a lower-paying job, providing grants for continuing education, and paying wage subsidies to employers who are willing to create more jobs – have “a close enough connection with the system’s basic objectives” and are therefore constitutionally valid extensions of the employment insurance system.

A Tax By Any Other Name…

The appellants challenged the constitutional validity of the Employment Insurance Act’s premium-setting mechanism, which authorized the government to adjust the EI premium rates each year. Section 66 of the Act authorized the Employment Insurance Commission to set the premium rates such that two objectives were satisfied:

(a) ensure that there will be enough revenue over a business cycle to pay the amounts authorized to be charged to the Employment Insurance Account; and
(b) maintain relatively stable rate levels throughout the business cycle.

Parliament later amended the Act to include section 66.1 (affecting the premium for years 2002 and 2003) and section 66.3 (affecting the premium for 2005). These provisions overrode section 66 and allowed the Governor in Council to change the premium rates directly, on the recommendation of the Cabinet. However, unlike s. 66, the new provisions did not enumerate any objectives that would guide how the premium rates should be set. The provisions’ failure to provide such guidelines, wrote the Court, effectively turned employment insurance payments into a general “payroll tax.”

Employment insurance has traditionally been viewed as a regulatory charge rather than a tax because premiums are levied to fund a specific regulatory scheme. The Court explored similar territory earlier this year in 620 Connaught Ltd. v. Canada 2008 SCC 7 (for which the case brief from TheCourt.ca can be read here). Referencing that case as well as Westbank First Nation v. British Columbia Hydro and Power Authority 1999 3 S.C.R. 134, Justice Lebel states that for there to be a regulatory charge, there must exist a regulatory scheme, the collection of a levy must be connected to that scheme, and the collected amounts must be used to fund the particular regulated activity.

Justice Lebel found that the second requirement of a regulatory charge – namely, the existence a relationship between the levy and the regulatory scheme – was not met by the premium-setting provisions of the Act. While section 66 sets out how the needs of the employment insurance system should guide the setting of the annual EI premium rate, the overriding sections 66.1 and 66.3 do not explain how EI revenues and EI expenditures are related. The new provisions allow the EI premium to be a tax for general purposes, managed by the Governor General in Council.

While section 53 of the Constitution allows Parliament to impose a tax, the creation of a tax must be explicit and “any taxing authority must be delegated expressly and unambiguously.” In this case, sections 66.1 and 66.3 of the Employment Insurance Act fail to state that the premium collection represents a tax and that taxing authority is delegated to the Governor General in Council. Therefore, the two impugned provisions (which affected EI premiums in 2002, 2003 and 2005) are unconstitutional.

Looking Forward: Consequences and Uncertainties

The Court’s broad reading of section 91(2A) of the Constitution may allow the federal government to breathe a sigh of relief as it now tries to deal with a $50 billion reserve. Since the reserve can only be used for employment insurance purposes, the Court’s open interpretation of the EI regime allows the government greater flexibility in how the money can be spent. Parliament may be tempted to use the funds for economic development initiatives that have only an indirect link to alleviating unemployment, as essentially any Parliamentary action that tethers unemployed persons to the labour market constitutes a valid EI expenditure.

The stickier issue concerns what will happen to the funds that were collected in 2002, 2003 and 2005. Since the premium-setting mechanism used at that time has been declared constitutionally invalid, the collection of funds during those years was unlawful. Justice Lebel’s reasons are silent on the subject of remedies, but he suspends the declaration of invalidity for twelve months, most likely to allow Parliament to come up with a solution. Given that the premium-setting mechanism was invalidated strictly because of inadequate language in ss. 66.1 and 66.3 of the Employment Insurance Act, I expect that Parliament will retroactively amend those provisions to adequately characterize the EI levy as a regulatory charge rather than a tax.