Labour market training of new Canadians and limitations to the intersectionality framework.

Hum, Derek ; Simpson, Wayne

ABSTRACT/RESUME

This essay considers the labour market integration of immigrants while distinguishing and assessing the importance of other identity markers including gender, visible minority status, and disability. Since much literature on the intersectionality of diversity markers is within a multiculturalism framework, we "map" the commonly mentioned diversity markers to mainstream economic research practices, pointing out limitations associated with the small case study as well as quantitative approaches. We sketch current economic understanding about labour market integration of Canadian immigrants and summarise recent research on their work-related training activities. We suggest an expanded framework to accommodate policy complications that result from Canada's federal-provincial division of powers. We conclude with remarks on the limitations as well as opportunities afforded by the intersectionality framework.

L'essai traite l'integration des immigrants dans le marche du travail tout en reconnaissant et en evaluant l'importance des autres marqueurs de l'identite, incluant le sexe, l'appartenance a une minorite visible et l'incapacite. Puisque la litterature sur l'intersectionalite des marqueurs de la diversite se situe dans un cadre de multiculturalisme, nous traons les marqueurs de la diversite les plus souvent mentionnes dans un contexte des pratiques de recherche economique traditionnelles, en soulignant les limites associees a une petite etude de cas ainsi qu'aux approches quantitatives. Nous dressons un portrait de la comprehension de la situation economique actuelle relative a l'integration des immigrants dans le marche du travail au Canada et resumons la recherche recente portant sur leurs activites de formation professionnelle. Nous proposons un cadre elargi pour composer avec les complications d'ordre politique resultant de la division des pouvoirs entre l'administration federale et les provinces du Canada. Nous concluons avec des commentaires sur les limites ainsi que les possibilites decoulant du cadre d'intersectionalite.

INTRODUCTION

Canada enjoys an advanced economy with progressive social programs, and it takes pride in welcoming immigrants. It is also a society that extols its multicultural character, even as it contends with its image of an officially bilingual country of two founding peoples and the perennial tensions that exist between English and French, East and West, the have and have-nots, and Ottawa and the provinces. Canada, then, is nothing if not diverse, and for this reason, it faces a daunting challenge to forge social cohesion and to promote inclusion, both civic and economic. Lambertus (2002:4) neatly summarises the policy evolution:

 Multiculturalism grew out of objections to the dualistic view of
 Canada in the 1969 Report of the Royal Commission on Bilingualism
 and Biculturalism. The resultant 1971 multicultural policy
 acknowledged Canada's ethnic diversity by recognising equal status
 of all cultures, and by protecting individual rights. In the 1980s,
 the focus shifted to managing diversity through institutional
 multiculturalism to provide employment equity and deal with systemic
 discrimination. Multiculturalism was incorporated into the 1985
 Canadian Charter of Human Rights and Freedoms, and the
 Multicultural Act [sic] was passed in 1988. For the past decade,
 multiculturalism remains committed to managing diversity through
 institutions, and more recently emphasizing citizenship
 participation of minority members, with an eye to enhancing society
 at large.

Although Canadians are multicultural and diverse, some are more multi- or more diverse than others. All of us have a "constellation of identities", including some "socially constructed" by others (Frideres, 2002:3). An individual's identity, it is argued, is defined by the social category to which one belongs. But some would add there are "fundamental traits" that determine the core identity of an individual; these may be the result of birth, history, or external factors and remain central in defining individual identity even under changing contexts. In short, there are many "identity markers" but these markers "intersect", thereby tempting us to consider how Canadian policies and programs can achieve inclusion and responsiveness to this intersecting nature of diversity. (1)

The markers of diversity enumerated in Canada typically include the customary ones of gender, official language spoken, visible minority status, age, immigrant status, Aboriginal status, disability status, as well as heritage language knowledge, ethnic ancestry, religious affiliation, region of residence, socio-economic status, racial category, sexual orientation, and marital situation, to name a few. One could stretch the list to include economic markers, such as occupation, sector of employment, or labour force status indicators such as employment, wage rate, and the like although this is uncommon. For the most part, discourse on diversity is largely absent from economists' conversations. (2)

This long list of identity markers speaks either to our limitless capacity in Canada to celebrate our individual differences, or else our collective need to manage our highly variegated multiculturalism (perhaps both!). Either way, a conundrum arises with respect to public policy. Intersectionality of diversity is predicated on the notion that simultaneous membership in two (or more) social categories can (may) have attendant consequences that are either unique to that combination of memberships, or alternatively, cannot be efficiently inferred from considering diversity markers either in isolation or in seriatim. This, if true, coupled with Canada's present population structure and changing mix of immigrant intake, raises the spectre of necessary policy adjustments to "make multicultural Canada work".

It is impossible to be anything but selective when considering intersecting diversity markers. Much writing on this issue posits such aims as promoting social inclusion, respecting differences to combat discrimination, or fostering citizenship participation within a paradigm that sees these diversity markers as fundamentally socially constructed. Our focus is somewhat different. Though also couched in terms of social equity, our attention is on the Canadian labour market and immigrants. One might even suggest that immigrants are the par excellence intersection of multicultural diversity. Immigrants comprise men, women, and children of different abilities and different religious beliefs; they settle in different parts of Canada, and possess different histories, socio-economic status, and official language fluency. Increasingly, immigrants to Canada are visible minorities. (3) For narrative convenience, we choose "new Canadians" as a core identity marker, and then further "intersect" immigrant status with such other diversity markers as gender, visible minority membership, age, etc.

Recently landed immigrants face many disadvantages in the Canadian labour market. The often-heard comment that immigrants are denied the same opportunities as other Canadians is sometimes countered by the claire that they do not have the same credentials as native-born Canadians, or alternatively, do not undertake the requisite training (human capital) because of pressures to maintain family consumption levels. But few studies exist beyond anecdotal evidence and small case studies. Furthermore, it is also important to note the special context and circumstances of the labour market integration of immigrants. Immigrants work, as well as age, after arrival in Canada, as do native-born Canadians. So how long must an immigrant remain in Canada before this identity marker becomes irrelevant to labour market expectation or outcomes? Do immigrants acquire human capital as easily as do native Canadians? Are there additional barriers or social capital deficits that afflict immigrants either temporarily or permanently? Are any of these circumstances amenable by public policy, and if so, by which level(s) of government?

This essay considers the training activities and labour market integration of immigrants while distinguishing and assessing the importance of other identity markers, including gender. Since much of the literature on the intersectionality of diversity markers is within a multiculturalist framework, the next section (Section 2) "maps" the commonly mentioned diversity markers to mainstream economic research practices, pointing out limitations associated with the small case study as well as quantitative approaches. Section 3 sketches current economic understanding about labour market assimilation by Canadian immigrants and summarises recent research on their work-related training activities. We propose an expanded framework in Section 4 to accommodate policy complications that result from Canada's federal-provincial division of powers. We conclude with remarks on the limitations as well as opportunities afforded by the intersectionality framework.

THE INTERSECTIONALITY PARADIGM AND MAINSTREAM ECONOMIC METHODS

Intersectionality refers to individuals with simultaneous membership in two or more social categories. The paradigm accordingly grants significance to the social category constructed by its own members, or in many instances, constructed by society itself. The group may often view itself as a community with a distinct culture, shared objectives, and specific aspirations as well as particular barriers and restraints. This contrasts with the dominant practice in mainstream economics in which the individual agent is centered. The basis for group construction in economics is usually homogeneity among "representative" individuals, rather than the "intersecting" diversity of individual characteristics. Heterogeneity among individuals often poses difficulties for empirical economic research. It is instructive to glimpse the relationship between the conceptual (non-economic) language of the intersectionality formulation and mainstream economic empirical research. Additionally, such an exercise highlights the contribution that different research programs can make, in addition to affording an appreciation of limitations. Despite their different language, there is much common ground.

The intersectionality paradigm uses some combination of diversity markers to define a group with a putatively shared objective or shared socio-economic condition. Economists, however, prefer to see these markers as simply individual characteristics or personal traits that are exogenous to the issue being examined, typically some economic outcome. If one is investigating employment earnings, for example, a marker (variable) such as religion or sexual orientation is regarded as "outside" the main determinants of earnings. Since there are so many marker/variables to consider, both in isolation and in combinations, the standard econometric approach is multivariate analysis methods. Intersectionality in the sense already noted--simultaneous membership in two (or more) social categories--is typically addressed formally by examining the so-called "interaction terms". (4) The detection of influence (if any) and measurement of the contribution arising from this "intersectionality" is thereby assessed in addition to whatever impact each marker/variable contributes individually. No simple model or single study can examine every possible influencing factor. There are, inevitably, special circumstances of history, or policy, or context to consider. For economists, markers of subsidiary interest to the main line of inquiry are included mainly for reasons of statistical control.

Writers within the post-structuralist framework frequently assert that identity markers are fluid and depend upon social construction or current negotiation. But despite shifting boundaries, the category itself will still retain currency, such as, for example, "female gender" or "Mainland Chinese". Pragmatic economists are content to adopt the social category label without worrying too much about changing interpretations that abound among separate sub-groups. Rather, economists worry more about distinguishing between those markers or tags that remain fixed (e.g., gender, ethnic heritage) and those that are more fluid, whereby individual behaviour can affect membership in the social category. For instance, one may become disabled through accident or advancing age, or cease to be illiterate through language training. The fluid markers, we shall see, have special significance for some lines of economic inquiry, since "exit" or voluntary entry is possible with certain social categories, such as disability (See Hum and Simpson, 2002a).

Another important distinction for econometricians is whether identity markers are observable or latent (unobservable). Some markers are easily observed (e.g., colour); others might be concealed by individual choice (e.g., sexual orientation) or semi-hidden (misdiagnosis of learning disability). Foreconomists, most troublesome are markers that are latent, such as individual motivation, ambition, attitudes, and the like that can play important roles in determining behavioural response.

One difficulty of a multicultural society in accounting for the myriad of diversity markers is the harsh reality of what may be termed "progression to the small", by which attempts to include more markers of diversity quickly reach a limit. This has implications for research as well as policy. The social circumstances of a small and narrowly defined group may not offer many lessons for a wider population. From a more practical level, a narrowly constituted group may be ineffective in marshalling voice for their circumstances or garnering political support to influence policy. The 'progression to the small" can be quite unforgiving.

Suppose the proportion of the population having identity marker of diversity 1 is P1, and that the proportion having identity marker of diversity 2 is P2, and so forth with P3 and P4 ... Then the population proportion simultaneously having diversity markers 1, 2, 3 ... will be the product of each individual population proportion; that is, P1 x P2 x P3 (if the markers are independent). This can be a very small number indeed. For example, consider gender as an identity marker of diversity, so the proportion of females in the population is roughly one half; that is, P1 = 0.5. Next, consider the population segment whose income is below the median, thus P2 is also one half, or 0.5. Finally, consider the population segment that is below the median age, so that P3 is 0.5. The "intersecting diversity" group of Canadians who are female with below median income and below median age is .5 x .5 x .5 = 1/8. In other words, after considering just three identity markers, in which the markers evenly divide the population, we obtain a group representing only 12.5% of the Canadian population. This is simply an example constructed for arithmetic convenience.

In fact, the identity markers of interest (besides gender) are typically very much lower in terms of population proportion (see Table 1). For instance, the proportion of the Canadian population that is disabled, or a member of a particular visible minority (say Chinese), or is an immigrant, or whose sexual orientation is homosexual, is much lower than the 50 percent in our example. Consequently, any attempt to consider more than three identity markers to constitute an intersectionality group is likely to be unproductive. One typically starts with a group that is small to begin with, for example, Canadians with disabilities (about 14% of the population). As soon as one incorporates the gender marker (P = 0.5) to examine, say, females with disabilities, the number drops by one half to 7 percent immediately. Finer grain contemplation of visible minority group membership, age bracket, region of residence, or language, shrinks the group rapidly.

The implication of the above discussion is as follows. Those who fix their gaze on intersecting diversity markers by considering too many social categories will face the unrelenting logic of progression to the small. (5) The risk is that the resulting group becomes irrelevant for analysis of political network building as well as program delivery. On the other hand, a multivariate analysis of intersectionality of the sort favoured by econometricians will need very large and complex data sets. Although census data provide coverage of the entire Canadian population and can be used to construct profiles of different diversity-combination groups, its drawback is that the response of interest for policy (e.g., wages or training) is often not available from the Census information. Panel data, on the other hand, are ideal for investigating processes through time, including changes in social category membership when individuals enter or exit particular groups. Additionally, this type of data can accommodate latent markers (see Hum and Simpson, 2000). However, this rich detail comes at a cost. Given the expense of gathering these types of data, sample sizes in panel data sets are often smaller than wished.

LABOUR MARKET PERFORMANCE OF IMMIGRANTS AND TRAINING

What is the state of knowledge about the labour market performance and identity markers of diversity? Consider work-related training activities of adult Canadians, focusing, first of all, on immigrant status as the main marker of interest. One is immediately urged to consider other identity markers, such as gender, age, language, region of residence, and the like, including, especially, visible minority membership, since a majority of Canada's new immigrants are non-white. Should visible minority membership or immigrant status be the core trait for labour market policy? Or something else?

Hum and Simpson (1999) examined wage offers for men and women using individual panel data to study the importance of the visible minority identity marker in determining labour market outcomes. They found immigrant status to be a more important diversity marker than visible minority status (except for black men). Hum and Simpson also included interaction terms between immigrant status and visible minority status and found it significant in revealing disadvantage for men (but less so for women). This therefore gives some credence to the notion that identity markers of diversity can have significance in particular combinations. Since immigration status appears to "dominate" in determining labour responses, we concentrate on this marker, allowing others such as gender to play secondary roles as necessary. (6)

Many studies report that immigrants face economic disadvantage in the Canadian labour market (Abbott and Beach, 1993; Baker and Benjamin, 1994 and 1997; Bloom, Grenier and Gunderson, 1995; Grant, 1999; Hum and Simpson, 1999; Li, 2000; McDonald and Worswick, 1998; Meng, 1987; Miller, 1992). Using cross-sectional data, these studies uniformly find an earnings gap (termed "the entry effect") at the time of landing relative to native-born Canadians with comparable characteristics. Many of these studies also find evidence of "economic assimilation" with time spent in Canada. More recently, however, Hum and Simpson (2000) find little evidence of wage convergence for immigrants when examining panel data from the Survey of Labour and Income Dynamics (SLID). They conclude that economic assimilation was largely absent or extremely slow in Canada during the 1990s. Hum and Simpson did find that acquiring additional training and work experience has a substantial effect on wage growth, thereby raising the further question regarding whether or not immigrants have the same training opportunities as the native-born.

We summarise preliminary research comparing immigrant and non-immigrant training patterns to illustrate the impact of several markers of diversity, bearing in mind that we have already pre-assigned priority to the markers of immigrant status and gender. Our "response" of interest is adult training (post formal schooling) for the workplace. (7)

We illustrate with a sample from the Adult Education and Training Survey (AETS) partitioned on the basis of immigrant status and the gender marker as shown in Table 1 (adapted from Hum and Simpson, 2003). We report training activity (defined as the duration of training in hours) for immigrant and native-born men and women on the basis of several personal characteristics including ethnic classifications (8) and disability status, plus a number of job characteristics and some core economic considerations such as education, age, and labour force attachment. Although we start with a large sample, Table 1 reveals that the intersection of only three categories (gender, immigrant status, and Black) results in very small sub-samples. For instance, only 54 of the full sample of 28,129 are male, black immigrants; this represents just 0.2 percent of the Canadian population. Nonetheless, through regression analysis we are able to estimate differences in training activity between diversity groups. We find that Blacks surfer a distinct disadvantage in the sense that they have significantly fewer annual hours of training that do non-Blacks, but that this disadvantage differs only across gender. While there is a distinctly greater disadvantage for black men than for black women, the disadvantage was no greater for immigrants than for the native-born: In fact, native-born black women have a greater training disadvantage vis-a-vis immigrant black women, once other personal and job-related factors have been considered. Similarly, those with disabilities (yet another identity marker) were at a disadvantage in training, and men with disabilities at a slightly greater disadvantage than women with disabilities, but immigrants actually fare better than their native-born counterparts.

The Canadian literature on immigrant economic assimilation is quite diverse despite its small size. Nevertheless, a few patterns emerge which we summarise as follows. Upon entry, male immigrants appear to work fewer hours at less secure jobs and for lower wages (compared to an equivalent native). This is especially true of later cohorts, most likely because they contain fewer immigrants from Britain, Europe, and the United States (and thus more visible minorities). On balance, the literature does indicate some assimilation along most dimensions, although the pace of assimilation is unclear. For the most part, female immigrants follow roughly the same pattern. However, women in two-immigrant marriages appear to work high initial hours in dead-end jobs, possibly to finance human capital investment activities by their husbands (see Hum and Simpson, 2000).

INTERSECTIONS OF DIVERSITY: OMISSIONS FOR POLICY STUDIES

How illuminating is a recasting of multiculturalism in terms of individual markers of diversity for addressing issues of public policy in Canada? The essentials of the intersectionality formulation of markers of diversity are captured in the Venn diagram of Figure 1. One dimension (horizontal axis) denotes a marker of diversity, and another (vertical axis) is labeled "response/goal" to denote the policy objective; for example, inclusion or citizenship or economic outcome. Separate circles depict individual diversity markers such as visible minority membership (e.g., Asian), gender (e.g., female), and immigrant status. Members of a particular social category, such as immigrants, are represented within the circle, and others (non-immigrants) lie outside the circle. The intersection or area common to all three circles (shaded in Figure 1) represents simultaneous membership in all three social categories. In this instance, the intersectionality group comprises females, Asian, and immigrants.

[FIGURE 1 OMITTED]

The premise of intersectionality is that this combination constitutes a unique group. Whatever the validity of the claire, the crucial point to note for policy design and program delivery is that initiatives targeted toward this putative "multiple jeopardy" group would then miss or exclude males (who may be immigrants and Asian), other Asians (who are not immigrants or female), and other immigrants (who are male and European or African or North American). Policies and programs directed at all females, or all immigrants, or all visible minority members would, of course, include the intersectionality group. Consequently, a strong onus will be on this particular intersectionality group to demonstrate the ineffective nature of the programs applied jointly, or the greater social (as opposed to recipient) benefits of such program targeting. This may be easy or difficult, depending on the particular group. Is a program restricted, say, to black immigrants to be preferred (by society as a whole) to separate general programs aimed at all immigrants and all visible minority groups? In sum, too literal and too aggressive a pursuit of intersectionality can result, ironically, in greater exclusion rather than more inclusion. Combined with the logic of the progression to the small, it is easy to see why this is so. But why cannot each group have its own special targeted program? There are many complications associated with program design and delivery.

Canada's federal constitution has implications for intersectionality as it relates to policy responsibility. This dimension appears largely absent in present-day discussions of intersectionality--a grave omission from a policy, public administration, and political perspective. In addition to the traditional wrestling over constitutional jurisdiction with respect to program responsibility and fiscal matters, additional issues should be noted for policy development and administration. In the situation of multiple diversity markers, one important requirement is that there must be as many policy instruments (programs) as there are diversity markers. For example, if one wishes to address issues pertaining to official language as well as immigrant economic integration, it will not be possible to do so with a single program for the simple reason that a program targeted primarily toward the concerns of immigrants may be ineffective in achieving the different aim of the official languages policy, and similarly, concentrating on promoting official languages may not pinpoint and address all the needs of immigrants. In the context of the markers of diversity, policies and programs from the multiculturalism policy sector alone will be ineffective. This is the well-known Tinbergen Principle (after the first Nobel Laureate in economics, Jan Tinbergen), whereby one must have at least as many policy instruments as there are policy targets.

Moreover, it is not simply a matter of having more policies and programs or requiring more government departments to adopt the intersectionality framework. There is also the matter of what is known as effective market classification, a principle due to another Nobel Laureate in economics, Canadian-born Robert Mundell. (9) Briefly, this principle demonstrates that program impact is maximised by assigning responsibility to the policy instrument or program that has the most direct, as opposed to indirect, influence on the goal. For example, one should not assign responsibility for employment equity to an agency devoted to protecting the environment, or even to local municipal government.

A further consideration relates to the nature of Canada's constitution. Immigrant matters happen to be a concurrent jurisdiction under the Canadian Constitution, so Ottawa and the provinces both have authority to act. The needs and preferences of each province respecting immigration intake must be coordinated and enforced by federal authority since immigrants, once landed, are free to move from one region to another, one city to another, one economic sector to another. The federal government must be responsible for setting an overall limit to the number and types of immigrants Canada accepts in the interests of all Canadians. On the other hand, labour market training is often best undertaken on a sub-national basis, thereby arguing for a provincial focus, as evidenced by the separate Labour Market Development Agreements (LMDA). What does this imply, then, for a coherent policy to assist new immigrants to acquire training for the Canadian labour market? Which order of government ought to deliver, or coordinate, programs directed toward particular intersectionality groups? In one form or another, this is the perennial tussle in Canada known as federal-provincial relations. (10)

The above points relate to policy responsibility and the design of program delivery. A comment can also be made from the perspective of the diversity groups themselves vis-a-vis government at both levels. Put bluntly, what is the best basis for securing voice, political attention, and policy action? Specifically, are the rights and entitlements of visible minorities better guaranteed through provincial auspices or federal policies? Should rights of women be considered independent of, or mainly alongside, the markers of age, or language, or sexual orientation? Are the aims of the disabled better achieved by disregarding the markers of region? These questions are partly rhetorical; they address the issues of "to whom" and "with which different others" ought a diversity-marked group to direct its political activities. There is no universal answer, but it is arguable that some groups may see their best course of action to lie in deliberately de-emphasising intersectionality to achieve as broad a voice as possible by limiting their goal to fostering advancement for their single diversity-marked group, or to direct their energy to a single (either federal or provincial) government order.

Our outlined extensions to the intersectionality framework can be glimpsed in Figure 2. Figure 2 incorporates Figure I (previously discussed) as a two-dimensional fiat plane, representing the relationship between identity markers and response. A third axis, denoting policy design and delivery, is now shown, and its space includes the two main orders of government in Canada. Lines emanating from each order of government and directed to different social categories, including possibly an intersectionality sector, are shown for illustrative purposes of the many possibilities of program delivery. Note especially, however, that responsibility for immigrants is shared between Canada and the provinces, and both orders of government may choose to deliver programs to the same group. Finally, local government, indirect agencies, NGOs, and the like are omitted from the diagram for simplification, although their roles may be important. However, it bears emphasising that the entire interacting relationship between orders of government and diversity-marked groups is "mediated" through the economy (in our specific application). The central point, then, is simply to warn that considering intersectionality of diversity in Canada in a policy context--even absent the complications of a complex and changing economic structure--must incorporate the fundamental fact of Canadian political and economic life, its federal constitutional nature, and what this means for promoting social equity, however viewed and pursued.

[FIGURE 2 OMITTED]

CONCLUSIONS AND CHALLENGES

The myriad of identity markers reminds us of the diverse and multicultural nature of Canada, and why the need to construct a tolerant civil society that respects differences should command our attention. But as the number of diversity markers increases, there is the danger that groups who claire "too special" a status by combining too many diversity markers will find their concerns ignored and their presence further marginalised. The possibility of multiplying social category enclaves is unattractive. At the same time, not all diversity markers may be equally relevant, and some markers may have greater importance in particular policy areas. Although virtually all diversity markers are, or ought to be, on a par with respect to such aims as social participation, the franchise, human rights, and the like, other more limited objectives justify ignoring some diversity markers as irrelevant. As well, the concerns of competing diversity groups will often conflict and need to be reconciled. Still other markers of diversity might claim greater relevance in particular policy areas, such as labour markets and economic outcomes.

The intersectionality formulation of Canada's multicultural character is useful. However, its limitations should be acknowledged. At present, it is too much the captive of simply enumerating and layering social categories without attention to alternative paradigms or realpolitik. There is also an insufficiently developed consensus on a hierarchy of markers and silence on the circumstance that might privilege some diversity markers over others in policy applications. There remains an inadequately developed confrontation of the federal-provincial dimensions of intersectionality and, most of all, a lacking in breadth in conversational partners. These are a few of the challenges posed by the very important issues of intersectionality and identity markers of diversity. Our progress in addressing them honestly will largely determine whether 'intersectionality of diversity" becomes a future guide-post for a multicultural Canada.

Table 1 Training Activity Among Canadian Men and Women by Immigrant
Status; Sample Size: 28,129

 Men (n = 12,423)
 Immigrants Canadian
 born
By Gender &
 Immigrant Status 1,468 10,955
Average Training Hours 23.4 29.8
Black (%) 54 (0.19) 44 (0.16)
Disabled (%) 191 (0.68) 1,677 (5.96)
Black Training Effect -43.4 -43.5
Disabled Training Effect -20.4 -24.9

 Women (n = 15,706)
 Immigrants Canadian
 born
By Gender &
 Immigrant Status 1,767 13,939
Average Training Hours 33.1 24.8
Black (%) 83 (0.30) 28 (0.10)
Disabled (%) 223 (0.79) 2,161 (7.68)
Black Training Effect -08.6 -24.0
Disabled Training Effect -16.5 -20.8

Source: 1998 Adult Education and Training Survey Master (Internal)
File; calculations by the authors

ACKNOWLEDGMENT

We thank the Prairie Centre of Excellence for Research on Immigration and Integration (PCERII) for financial support. The views expressed are solely those of the authors.

NOTES

(1.) This paraphrases the statement of purpose of the Intersections of Diversity Seminar held in Niagara Falls, April 2003, sponsored jointly by the Association for Canadian Studies, Canadian Heritage, and the Metropolis Project.

(2.) As anecdotal (rather than compelling) evidence, the reviews (authors) commissioned for the Intersections of Diversity seminar covered the following "markers": Aboriginal peoples (Lambertus), Age (Fieras), Disability (Stienstra), Gender (Cohen), Immigration (Frideres), Official Languages (Quell), Region (Nurse), Sexual Orientation (van der Meide), and Socio-economic Status (Kazemipur). All the authors argue, with varying degrees of insistence, that groups or society socially construct the markers of diversity. None of the authors is an economist and it may simply be that no economist was asked to write. Nothing sinister should be taken from this description aside from the fact that economists are not prominent voices in this discourse. Sec Hum (1995) for a discussion of how multiculturalism can affect an economy's structure.

(3.) Two of every three immigrants to Canada arc visible minorities. The Employment Equity Act defines visible minorities as "persons, other than Aboriginal peoples, who are non-Caucasian in race or non-white in colour." The regulations to the Act specify the following groups as visible minorities: Chinese, South Asians, Blacks, Arabs and West Asians, Filipinos, Southeast Asians, Latin Americans, Japanese, Koreans, and Pacific Islanders.

(4.) These will include both "slopes" and "intercepts" in a linear model, as well as other terms in more complicated functional specifications.

(5.) It is also the case that not just any combination of diversity markers will resonate with the public. Further, combinations comprising too many markers will have no mass; e.g., it may be hard to find many individuals in the social category of bilingual, non-immigrant, Chinese, male, disabled, atheist from the Maritimes.

(6.) Gender is an important identity marker or "fundamental trait" in labour economics. all economists acknowledge that labour supply responses, as well as market opportunities, differ for men and women, so no special pleading is ever necessary to prompt labour economists to consider gender as a diversity marker in research or policy concerning labour markets. We simply choose to focus on immigrants for narrative presentation purposes.

(7.) We employ data from the master (non-public) files of the 1998 Adult Education and Training Survey (AETS). Since the 1998 AETS distinguished immigrants and the native-born for the first time, we are able to compare the training activities of immigrants to comparable Canadians (see Hum and Simpson (2002c) for details). Private sector workplace training is examined in Hum and Simpson (1996), and government-sponsored training programs are assessed in Hum and Simpson (2002b)

(8.) Ethnic categories are limited and incomplete in the AETS, but we were able to identify categories such as Blacks, which we use for illustrative purposes here.

(9.) A large economic literature exists and continues on these topics. See the relevant essays in Wahid (2002) for non-technical explanations of the seminal Nobel contributions of Tinbergen and Mundell.

(10.) The discussion in the main text is deliberately kept general but the federal-provincial aspect is especially noteworthy in our example respecting labour markets, since much of the program delivery is sub-national or provincial. Occupational Health and Workers' Compensation are administered provincially; minimum wage rates are set by provinces; union certification and regulation is at the provincial level, as is governance of the professional self-regulating bodies (lawyers, doctors, etc.) that assess credentials; and employment equity legislation, to name a few. For our particular focus, education and training is delivered at the provincial level. Immigration is, of course, a shared jurisdiction.

REFERENCES

Abbott, M. and C. Beach. (1993). "Immigrant earnings differentials and birth-year effects for men in Canada: Post-war--1972." Canadian Journal of Economics 26(3): 505-524.

Baker, M. and D. Benjamin. (1994). "The performance of immigrants in the Canadian labour market." Journal of Labor Economics 12(3): 369-405.

Baker, M. and D. Benjamin. (1997). "The role of the family in immigrants' labor-market activity: An evaluation of alternative explanations." American Economic Review 87(4): 705-727.

Beach, C. and C. Worswick. (1993). "Is there a double-negative effect on the earnings of immigrant women?" Canadian Public Policy/Analyse de Politique XIX(1): 36-53.

Bloom, D. E., G. Grenier, et al. (1995). "The changing labour market position of Canadian immigrants." Canadian Journal of Economics XXVIII(4b): 987-1005.

Frideres, J. (2002). "Immigrants, integration and the intersection of identities." Paper commissioned for Seminar on Intersections of Diversity.

Grant, M. (1999). "Evidence of new immigrant assimilation in Canada." Canadian Journal of Economics 32(4): 930-955.

Hum, D. (1995). "Multicultural society and economic structure." In H. Braun and W. Klooss (eds.), Multiculturalism in North America and Europe: Social practices--Literary visions, Trier, Germany: Wissenschaftlicher Verlag Trier.

Hum, D. and W. Simpson. (1996). Maintaining a competitive workforce: Employer-based training in Canada. Montreal: Institute for Research in Public Policy.

Hum, D. and W. Simpson. (1999). "Wage opportunities for visible minorities in Canada." Canadian Public Policy 25(3): 379-394.

Hum, D. and W. Simpson. (2000). "Closing the wage gap: Economic assimilation of Canadian immigrants reconsidered." Journal of International Migration and Integration 1 (4): 427-441.

Hum, D. and W. Simpson. (2002a). "Disability onsets among aging Canadians: Evidence from panel data." Canadian Journal on Aging 21(1): 117-136.

Hum, D. and W. Simpson. (2002b). "Public training programs in Canada: A meta-evaluation." Canadian Journal of Program Evaluation 17(1): 119-138.

Hum, D. and W. Simpson. (2003). "Job-related training activity by immigrants to Canada." Canadian Public Policy 29(4): 1-22.

Lambertus, S. (2002). "Canada's Aboriginal peoples and intersecting identity markers: Research and policy implications for multiculturalism." Paper commissioned for Seminar on Intersections of Diversity.

Li, P. S. (2000). "Earnings disparities between immigrants and native-born Canadians." Canadian Review of Anthropology and Sociology 37(3): 289-311.

McDonald, J. and C. Worswick. (1998). "The earnings of immigrant men in Canada: Job tenure, cohort, and macroeconomic conditions." Industrial and Labor Relations Review 51(3): 465-482.

Meng, R. (1987). "The earnings of Canadian immigrant and native-born males." Applied Economics 19:1107-1119.

Miller, P. (1992). "The earnings of Asian male immigrants in the Canadian labor market." International Migration Review 26(4): 1222-1247.

Wahid, A., ed. (2002). Frontiers of economics: Nobel laureates of the twentieth century. Westport, CN: Greenwood Press. 18-27; 381-389.