Home Research Projects The Vulnerable Worker Publications Vosko - Confronting the Norm Part 3

Research Projects

The Vulnerable Worker

Publications

Confronting the Norm: Gender and the International Regulation of Precarious Work

<< Back to Table of Contents

PART FOUR: Limiting Precarious Work?  Alternative Approaches >

PART THREE: A Portrait of Precarious Work in Australia, Canada, and the United States

To assess the viability of new and emerging international labour standards on part-time work, home work, and the scope of the employment relationship, it is necessary to examine labour force trends at the country-level.  To this end, this section develops a portrait of precarious work in Australia, Canada, and the United States.[20]  In each of these countries, precarious work is characterized by job insecurity, low wages, limited social benefits and statutory entitlements, and a lack of control over the labour process.  In each instance, it is also gendered.  Yet there are significant differences.

In Australia, precarious work pivots mainly on time-based deviation from the standard employment relationship.  It takes sharp expression within a subcategory of part-time employment that falls outside the shorter-hours employment norm envisioned under the Convention on Part-Time Work – part-time ongoing casual.  A disproportionate number of Australian women, many of whom have young children, fall into this category.  In Canada, deviation from the standard employment relationship on the basis of status grew in significance in the 1980s and 1990s.  In this context, a sizeable proportion of full-time solo self-employment not only lacks qualities associated with entrepreneurship, a reasonable measure of genuine commercial activity, but is precarious regardless of whether it is disguised.  A large segment of Canadian women in full-time solo self-employment, many of whom engage in self-employment to fulfill care-giving responsibilities, also have very low-incomes and lack benefits and/or independent access to benefits.  By its very definition, an international labour standard on the scope of the employment relationship focused on disguised employment and objectively ambiguous situations could never offer a model for extending protections to this group.  In comparison to Australia and Canada, full-time permanent employment still dominates overwhelmingly in the United States.  Here, while there is certainly some deviation from the standard employment relationship, which also tends to involve precarious work, it is the erosion of the employment norm itself that is profound.  Among full-time permanent employees, conditions of work are deteriorating.  The effects of this process of erosion are highly gendered as many American women, especially single women with young children, hold full-time permanent jobs that are highly precarious.

I.     Measuring the Standard Employment Relationship

Full-time permanent employment is the best statistical indicator for the standard employment relationship.  But data available for Australia, Canada, and the United States over time allow only for the consideration of two facets of this norm – employee status and full-time work.  Time series data are not available on a third central facet (i.e., permanency). 

Figure 1 depicts full-time employment in these countries in the late 20^th century.  Instability in full-time employment dates to the early 1970s in each of these countries.  Yet patterns of instability differed markedly in Australia, Canada, and the United States between 1968 and 2002.  Full-time employment slipped from 90 to 71 percent of total employment in Australia in this period.  Trends were similar, although less dramatic, in Canada, where it declined from 90 to 81 percent of total employment.  The United States, in turn, experienced a modest decline in the 1970s.  However, full-time employment stabilized at a relatively high level in the early 1980s – and it stood at approximately 83 percent of total employment in 2002.

Click on the picture to enlarge

Permanency is an important feature of the standard employment relationship.  Table 1 provides a snapshot of the standard employment relationship and its deviations that includes this central feature.  It breaks down total employment in Australia, Canada, and the United States for 2000, using a new approach for statistical conceptualization and measurement.[21]  To capture the diversity in the forms of employment differing from the standard employment relationship, Table 1 first elevates employment status by differentiating between paid employees and the self-employed.  It then divides the self-employed into those without employees (solo self-employed) and those who employ others (employers).  In parallel, it separates permanent and temporary employees.  Finally, it splits each subgroup of employees and self-employed people by full-time and part-time status. 

This approach to breaking down total employment at a country level links forms of employment deviating from the standard employment relationship to dimensions of precarious employment.  According to Rodgers (1989, 3-5), four dimensions are central to establishing whether a job is precarious: control over the labour process; degree of certainty of continuing employment; degree of regulatory protection or whether the worker has access to social and labour protections; and, income level.  These dimensions are the point of departure for a growing body of scholarship on precarious work among employees in both Canada and Australia (Armstrong and Laxer forthcoming; Campbell and Burgess 1998; Vosko, Zukewich and Cranford 2003) and also among the solo self-employed (Vosko and Zukewich forthcoming).

The first order of distinction in Table 1 (i.e., between employees and the self-employed) is central to workers' capacity to exercise control over the labour process and their degree of regulatory protection.  The compendium of standards comprising the ILC extends labour and social protections most fully to workers with an identifiable employment relationship.  Correspondingly, in Australia, Canada, and the United States, few self-employed workers have access to collective representation through a union (Cranford, Fudge, Tucker and Vosko forthcoming; Cobble and Vosko 2000; Clayton and Mitchell 1999; Piore 2002).

 The second order of distinction addresses degree of certainty of continuing employment by grouping employees according to job permanency and by distinguishing between the solo self-employed and employers.  In each of the three countries, the solo self-employed are more vulnerable to uncertainty than employers (Fudge, Tucker and Vosko 2002; Hyde 2000; O'Donnell 2004).  Amongst employees, the category “permanent” signifies durability in the employment relationship, indicated normally by an indefinite contract of employment, while the category “temporary” approximates uncertainty.  In Canada and the United States, all forms of temporary employment fall, as one would expect, within the category “temporary.”[22]  The terminology is employed somewhat differently in Australia, however, where this category includes all casual employment[23] as well as employment on a fixed-term contract or paid by an agency.  As Anthony O'Donnell (2004, 18) argues, there is a strong case to be made for “aggregating those jobs which, regardless of their extended tenure (or prospect for extended tenure), grant relatively unfettered power to the employer to terminate by virtue of their regulatory designation.”[24]  Grouping together all casual employment, employment on a fixed-term contract, and employment paid by an agency makes it possible to compare Australia with Canada and the United States.[25] 

The third and final order of distinction in Table 1 also addresses access to social and regulatory protection in these countries since eligibility for and/or level of certain social benefits is pegged to hours of work.

In numeric terms, the standard employment relationship is quite fragile – especially in Australia and Canada (Table 1).  Only 48 and 63 percent of people in the labour force in Australia and Canada respectively held full-time permanent jobs in 2000 – significantly lower than in the United States, where 72 percent held full-time permanent jobs that year. 

More than half of total employment in Australia and over one-third in Canada differs from the employment norm.  Yet deviation takes distinct forms in Australia and Canada:  in Australia, temporary part-time employment is dominant (Figure 2a).  In Canada, solo self employment grew markedly in the 1980s and 1990s and full-time solo self-employment represents the largest subset of self-employment (Figure 2b).  Conversely, in the United States, almost three-quarters of total employment is both full-time and permanent (Figure 2c). 

Click on the picture to enlarge

Click on the picture to enlarge

Click on the picture to enlarge

Given these differences, it is worth exploring the character and quality of part-time work in Australia, solo self-employment in Canada, and full-time permanent employment in the United States.  As a backdrop for this discussion, Table 2 sets out statistical indicators used to depict dominant forms of deviation in Australia and Canada and erosion in the United States as well as dimensions of precarious employment and gender relations for each context.

Table 2: The Indicators - Australia, Canada and the United States
Australia	Indicator
Deviation from Standard Employment Relationship
Time - Testing the viability of the shorter hours employment norm	Part-time Ongoing Casual
Dimensions of Precarious Employment
Certainty	Permanent/Temporary
Income Level	Average Weekly Wages
Gender Relations	Marital Status
	Age of Youngest Child
	Industrial and Occupational Segregation
Canada	Indicator
Deviation from Standard Employment Relationship
Status - Exploring the efforts to expand the scope of the employment relationship	Full-time Solo Self-employment
Dimensions of Precarious Employment
Regulatory Protection	Benefit Coverage (Health, Dental and Disability)
Income Level	Average Annual Income (Four Income Groups)
Gender Relations	Reason for Self-employment
	Source of Benefit Coverage
United States	Indicator
Erosion of Standard Employment Relationship
Employment Norm - Examining the substance of the "standard employment relationship"	Full-time Permanent Employment
Dimensions of Precarious Employment
Control	Union Membership
Certainty	Permanent/Temporary
Regulatory Protection	Benefit Coverage (Medical, Dental and Vision)
Income Level	Average Weekly and Hourly Wages
Gender Relations	Marital Status
	Presence of Young Child

II.    Time- and Status-Based Deviation from the Employment Norm

The international regulation of precarious work seeks to stretch the standard employment relationship; it aims to normalize deviations based on time, place, and status through the adoption of an equal treatment approach.  To curb precarious work, new and emerging international labour standards attempt to expand traditional conceptions of normal working hours and place of work and to restore the employment relationship as the fulcrum of labour law and policy.  Does the tenor of this approach offer potential in Australia and Canada?  Specifically, does the Convention on Part-Time Work provide a model capable of limiting precarious part-time work in the Australia?  How might a recommendation on the scope of the employment relationship focused on limiting disguised employment relationships and addressing ambiguous situations affect self-employed workers in precarious situations in Canada?

A. Part-Time Work in Australia

Australia is a good test-case for the Convention on Part-Time Work, specifically its strategy of fostering equal treatment by redrawing the temporal boundaries of the employment norm.

Part-time work is far more prevalent in Australia than in Canada and the United States – and it is highly gendered too: 46 percent of women work part-time in Australia as opposed to 28 and 26 percent in Canada and the United States respectively (Figure 3). The magnitude of part-time work for women reflects historical patterns in Australia, where women's low level of participation in full-time employment has remained relatively constant since the 1930s in sharp contrast to Canada and the United States.[26] 

Click on the picture to enlarge

The composition of part-time employment al so distinguishes Australia from its Canadian and American counterparts.  In Australia, part-time employment is more than twice as likely to be temporary than in Canada and over five times more likely to be temporary than in the United States (Figure 4). 

Click on the picture to enlarge

Breaking part-time work down into its different forms provides greater insight into its core components and provides greater insight into time based-deviations from the standard employment relationship in Australia.  Accordingly, Figure 5 divides part-time work into its six dominant forms: permanent paid employment; ongoing casual paid employment; fixed-term[27] casual paid employment; fixed-term non-casual paid employment; solo self-employment; and, employer self-employment.  It illustrates that a relatively small proportion of part-time workers conform to the ideal type of the shorter-hours employment norm in Australia.  As a percentage of total part-time employment, just 27 percent of all those who work part-time are permanent employees.  The remaining 73 percent are either employed on a casual or a fixed-term basis (55 percent) or self-employed (18 percent). 

Click on the picture to enlarge

Interpreted in the broadest sense, the Convention on Part-Time Work covers all part-time wage earners, excluding the self-employed.  However, it only applies fully to part-time permanent employees since those engaged on temporary or casual bases may be excluded (ILO 1994, Articles 1a, 3.1).  These permissible exclusions are highly significant in the Australian case:  even among all part-time employees, only 33 percent fall within the shorter-hours employment norm (Figure 6).  Figure 6 depicts part-time employees as a group, illustrating that 61 percent are both ongoing and casual, while the remaining 6 percent are fixed-term casual (4 percent) and non-casual (2 percent).  A disproportionate percentage of part-time employees fall into the ongoing casual category,[28] not surprisingly since casual employment accounted for two-thirds of the growth of total employment between 1990 and 2001 (Watson, Buchanan, Campbell and Briggs 2003).

Click on the picture to enlarge

Given its magnitude, the meaning of casual employment in Australia and the significance of the ongoing casual category for part-time workers merit examination. In Australia, casuals normally work under a contract of employment but they lack the full range of labour and social protections enjoyed by permanent employees.  Casuals do not commonly “accrue those entitlements that assure some level of income security: sick leave, annual leave, severance pay, maternity leave and, in the majority of cases, protection against unfair dismissal” (O'Donnell 2004, 26).[29]  Under the awards system, casuals are also generally defined “as people that are paid as such” (Campbell and Burgess 2001, 177);[30] this language aims to convey that casuals are paid at the end of each engagement, although this is not always the case.  Unique to this form of employment, casuals also receive a form of precarity pay or “casual loading,” normally around 20 percent, or payments in lieu of entitlements (Owens 2001).   Many casuals rely on this compensatory premium because of their low wages.  Casuals also have lesser protection against unfair dismissal than their permanent counterparts under the Workplace Relations Act (1996); whereas the probationary period is normally three months, they must be employed for twelve months before they are protected from unfair dismissal (Regulation 30 B (1) (d)).  The increase in casual employment in the 1990s is attributable partly to provisions under the Act.  Under the Workplace Relations Act, the power of the Australian Industrial Relations Commission to make or vary industrial awards is curtailed –  the Commission can no longer limit the number or proportion of employees that an employer may employ in “a particular type of employment” (s. 89 A (4)).   Unpaid parental leave is also accessible to a very limited group of casuals.  Only casuals covered by certain federal awards who have worked for an employer on “a regular and systematic basis” over a period of at least twelve months and who have “a reasonable expectation of ongoing employment on that basis” may have access to this statutory entitlement (s. 53 (1),(2) and s. 57).[31] 

Casuals may be employed under various types of contracts.  There are those engaged on a “once-off” basis for whom a host of social and labour protections simply do not apply (Stewart 1992).  For others, with fixed-term contracts, access to such protections depends on the length of the term.  And there are casuals with ongoing employment relationships that should have extensive labour and social protections, such as protection against unfair dismissal.  For this group, however, enforcement is a major obstacle in gaining access to protection against unfair dismissal – even after twelve months of continuous service.  As Joo-Cheong Tham asserts (2003, 8): “more often than not, the employer and the casual employee would not have expressly adverted to this question [i.e., whether their relationship is ongoing].  In these circumstances, it becomes a nice question of fact where the casual employee is engaged on an ongoing contract.”  These ongoing casuals do not have leave entitlements, they have lesser protections against unfair dismissal than their permanent counterparts, and they have historically lacked entitlements to parental leave.  And yet they are engaged in a continuous way.  As Rosemary Owens (2001, 18) contends, ongoing casual employees represent a “distinct class” in Australian society. 

In distinguishing between casual and permanent part-time employees, the broad objective is to develop a deeper understanding of the socioeconomic situation of part-time employees that are precarious by definition – i.e., those that lack leave entitlements and generally have lesser protections than full-time permanent employees. The narrower aim is to isolate this “distinct class” – i.e., part-time casuals whose work is ongoing – and gain greater insight into its size and characteristics, as well as its composition.

Part-time ongoing casuals are by far the largest group of part-time employees in Australia.  And women are over-represented in this group.  Out of all women part-time employees, 25 percent are ongoing casual in contrast to just 11 percent of all men (Figure 7).  Furthermore, 68 percent of part-time ongoing casuals are women as opposed to 32 percent of men (Figure 8). According to Pocock et al (2004, 20), this gender imbalance emerged partly because of the peculiar and longstanding “conjunction of permissive regulation of casual work with strict regulation of part-time work.” 

Click on the picture to enlarge

Click on the picture to enlarge

For many workers seeking part-time jobs in Australia, the only option is to accept casual positions since permanent part-time employment is so undeveloped, except in the public sector.  Trends by industry and occupation attest to this claim.  They show how regulatory practices in these industries shape the gendered character of the part-time ongoing casual category.  Awards in key female-dominated occupations and industries rarely provide scope for expanding the number of part-time permanent employees whereas awards in male-dominated industries and occupations routinely impose rigid restrictions on the use of casuals.[32]  As Poccock et al (2004, 24) show, referring to a key application for a change in an award,[33] half of the eighty-six awards in manufacturing set maximum periods of engagements for casuals of two to four weeks and over two-thirds set eight-week maximums.  Few awards in female-dominated occupations and industries set such limits.

Industrial and occupational segregation shape gendered patterns of precarious part-time work in Australia.  By industry, accommodation, cafes and restaurants, followed by retail trade and cultural and recreational services have the highest levels of part-time ongoing casual employees.  Furthermore, in the two industries with the highest levels, a majority of women employees hold part-time ongoing casual positions (Figure 9).  Patterns are similar by occupation, although women in clerical, sales and service occupations, a highly female dominated group, are particularly likely to be engaged on a part-time ongoing casual basis (Figure 10).  Key awards have cultivated this situation.  They have contributed to the magnitude of the part-time ongoing casual category within this occupational group by restricting access to permanent part-time employment. Under the South Australian Clerks Award, for example, it was only in 1988 that it became possible to work part-time hours other than as a casual (Pocock et al 2004, 20).  In professional occupations, part-time ongoing casual jobs are also quite common, especially among women, who are three times more likely than men to hold these jobs.  The prevalence of part-time ongoing casual employment among women is characterized not only by continuity (i.e., traditional practices in female-dominated occupations) but also change (i.e., the use of this category to casualize employment in various occupational groups).

Click on the picture to enlarge


Click on the picture to enlarge

Which women are concentrated in part-time work in Australia and, more specifically, which women are ongoing casuals?    A higher percentage of women living with a partner than single women and a higher percentage of women with young children than women without young children are part-time employees (Figure 11).  And 83 percent of women with young children, in contrast to just 10 percent of their male counterparts, are part-time employees (Figure 12).  These patterns mirror trends in Canada and the United States, where part-time work is primarily the domain of women with young children and some (primarily young) men pursuing their education (Applebaum 2002a; Armstrong and Armstrong 1994; Duffy and Pupo 1992; Rosenfeld 2001).  What makes Australia unique is its large proportion of women employees with young children that are part-time ongoing casual.  Out of all women employees with young children, 32 percent fall in this group in contrast to just 5 percent of men (Figure 13).

Click on the picture to enlarge

Click on the picture to enlarge

Click on the picture to enlarge

The structure of regulatory protection in Australia means that part-time ongoing casuals are precarious almost by definition; they lack leave entitlements and have limited access to labour and social protections.  Data on sex-based income gaps strengthen this claim and illustrate that it is deeply gendered. Comparing women's weekly income relative to men's, part-time ongoing casuals have the largest sex-differential among all part-time employees.  On a weekly basis, women part-time ongoing casuals earn 88.2 percent of their male counterparts.  The income gap narrows among those that are fixed-term and it is non-existent among those that are permanent, groups historically more successful in closing the gap due to Australia's strict regulation of certain forms of employment and the accommodating approach to others (Figure 14). 

Click on the picture to enlarge

1. The Shorter-Hours Employment Norm in Context

The Australian case highlights the limits of the ILC approach to curbing precarious work, focused as it is on limiting time-based deviations from the standard employment relationship.  In its aim to incorporate more “regular” part-time workers within the employment norm, the Convention on Part-Time Work is concerned to bring those forms deviating least from the standard employment relationship into the norm.  If the ILC model were applied in Australia, the primary group served would be part-time permanent employees.  While the shorter-hours standard employment relationship has the potential to improve the situation of this group, this model neglects not only the most sizeable segment of part-time employees but those that are worst off.  Among part-time employees, it is part-time ongoing casuals whose work is especially precarious along multiple dimensions.  Part-time ongoing casuals are far more likely to confront low income and they have the largest sex-based income differential among part-time employees.  This “distinct class,” moreover, is highly female-dominated, a feature exacerbated by industrial and occupational segregation.  And it is composed not only of many married women (and few men), but many women with young children. Set against the dominant model for re-regulation, the Australian case highlights the importance of minimum standards since their decline and/or absence in this context has cultivated an expansion of the part-time ongoing casual category, with serious consequences for women, especially those with young children.  Women with young children are highly vulnerable to becoming part-time ongoing casuals because of the assumption, at a policy level, that they are responsible for unpaid care-giving: it is these workers that require minimum standards most. 

Some changes are afoot in Australia – changes oriented towards remedying the situation of ongoing part-time casuals and fostering permanent part-time employment.  Specifically, there have been attempts to foster conversion to permanent full- and part-time employment among ongoing casuals, principally at the state level, in female-dominated occupations.  The chief example is the South Australia Clerks Award, which grants ongoing casuals with 12 months service the right to request to become permanent, a request that the employer cannot refuse on unreasonable grounds.  Yet this approach has definitive limits.  As Owens (2002) argues, instead of extending the safety net to casuals automatically, workers must elect to convert, a “choice” that obscures power imbalances between employees and employers.  The precarious situation of part-time ongoing casuals makes them highly vulnerable to reprisal from employers; this vulnerability, combined with the fact that many part-time ongoing casuals are dependent upon the premium provided through casual loadings, contributes to the lack of test-cases to date (Pocock et al 2004, 43).

There have also been limited advances in compensating casuals for their precarious situation, although most have been industry-specific and are not entrenched in all awards or industrial law (Pocock et al 2004, 39).  The backdrop to these minimal improvements, moreover, is a general retreat from what Pocock et al (2004, 37) characterize as a “limitation approach,” whereby casuals are not to be engaged beyond a specified time period or numerical limits or ratios are imposed.  This retreat reflects the growth of enterprise-level bargaining, and the general withdrawal of award restrictions, promoted under the Workplace Relations Act (Clayton and Mitchell 1999).  In parallel, there is a growing movement to increase the scope for part-time permanent employment.  For example, the Workplace Relations Act introduces the concept of regularity for part-time employees and more awards reflect this policy direction; casting this movement positively, some analysts argue that it should enable employees to achieve a better work-life balance (Buchanan and Thornthwaite 2001, 32). 

There are two prongs to current policy proposals aimed broadly at limiting part-time ongoing casual employment: the first involves limiting casual work to genuinely irregular, intermittent or on-call work.  Behind this policy emphasis is the perception that regulators have lost sight of the true meaning of casual.  As the argument goes, “casual” has not been defined and interpreted properly and, consequently, there is a regulatory gap.  There is, however, considerable debate over the nature and size of this gap and the magnitude of its consequences for Australian workers; there are those that acknowledge this gap and call for a commonsense understanding of the phenomenon, while minimizing the severity of the regulatory gap itself (see for example: Murtough and Waite 2000a and b), and there are those that reject the “vernacular” (O'Donnell 2004) understanding of casual and call, in contrast, for more dramatic changes in regulatory protection  (see also Owens 2001; Tham 2003).  For the latter group, the proposed remedy is sound.  It aims to “prevent ‘casual' employment status from being abused and to ensure that – as in other OECD countries – it is confined to its proper place as just a minor component in the range of employment forms” (Pocock et al 2004, 47). 

Informed by a similar rationale, the second prong involves cultivating a shorter-hours employment norm akin to the ILC model and thereby designed to coexist with, rather than displace, the standard employment relationship.  Support for this model is evident in policy areas, such as parental leave, where there are attempts to extend entitlements to ongoing casual part-time workers on the basis of equal treatment.  These proposals cast part-time work positively, as enhancing social and economic objectives and, in particular, helping “people strike a better balance between work and life as they navigate transitions to and from work and education, work and family formation, spells of unemployment and as they prepare for retirement” (Pocock et al 2004, 47).  Their aim is to construct “a new deal for part-time workers, to improve the quality of their jobs and ensure access to part-time work for parents who need it” (Buchanan and Thornthwaite 2001, 2).  They understand the “Australian problem” as a problem of poor quality part-time jobs.  The idea is to improve part-time jobs by decasualizing those that are ongoing.  The main criticisms of a shorter-hours employment norm targeted mainly to women with family responsibilities relate to its potential to perpetuate a low-wage segment, mainly in service industries.  For example, both Ann Junor (1998) and Belinda Probert (1995, 1997) argue that part-time permanent employment can enable employers to avoid overtime because employees that are part-time may be asked to work up to full-time without supplemental pay.  These authors also point to the limited prospects for vertical mobility and the potential for forms of shift work and variable hours that suit the needs of employers but not workers (Probert 1995, 40-42). 

The fundamental problem with both prongs is that they are cast in gendered terms in that they are designed to enable women to “balance” care-giving and paid employment.  Furthermore, neither takes the limits of equal treatment seriously – that is, the deficiencies of dominant definitions of equivalency (e.g., graduated protection).  In Australia, even those part-time casuals that successfully convert to permanent status receive only pro rata access to the labour and social protections available to full-time permanent employees.

B.     Self-Employment in Canada[34]

Canada is a suitable case study for evaluating the international labour standard on the scope of the employment relationship, which is poised to provide guidelines for extending labour protections to dependent workers in disguised employment relationships and clarifying situations that are ambiguous.  Standing at 15 percent of total employment in 2000, Canada has a relatively high proportion of self-employment among OECD countries such that it sits at a midpoint between Australia and the United States (Figure 15; see also Figures 2a-c).  From the late 1970s to 1998, self-employment[35] grew at a faster rate than paid employment.  Still, self-employment in Canada is quite varied, cutting across industry and occupation; it may involve employing others or working solo; and, it may be full-time time or part-time.  The significance of this heterogeneity is that entrepreneurship and self-employment do not necessarily coincide (Fudge, Tucker and Vosko 2002, 16; see also: Dale 1991; Rainbird 1991).  Self-employment is traditionally associated with ownership, autonomy, and control over production, key features of entrepreneurship (Eardley and Cordon 1996, 13).  At a regulatory level, it is cast as a form of independent contracting and thereby outside the scope of labour protection.  Yet many self-employed people resemble employees more than entrepreneurs (Fudge, Tucker and Vosko 2003b).  This is especially true for those that do not employ others – especially women.  Thus, the foremost feature differentiating the self-employed is employer status, best measured by the distinction between the solo self-employed and self-employed employers (Fudge Tucker and Vosko 2002, 99).  Self-employed employers are more likely than their solo counterparts to possess the skills necessary to engage in longer-term business planning associated with entrepreneurship (Fudge, Tucker and Vosko 2002, 21).[36]  The solo self-employed, in contrast, are more likely to receive some kind of support from their clients than their employer counterparts[37] and many have very few clients.[38]

Click on the picture to enlarge

Solo self-employment drove the growth in self-employment in the 1990s and has since stabilized at relatively high levels (Figure 16).  Legal and statistical categories are somewhat incongruent (Fudge, Tucker and Vosko 2002).  The statistical measure of solo self-employment is very broad.  It covers self-employed contractors that are not dependent on a single employer, independent contractors that may work directly for a contractor, and disguised employees who, in the narrow definition adopted by the OECD, may “work for just one company, and whose status may be little more than a device to reduce total taxes paid by the firms and the workers involved” (OECD 2000, Chapter 5, 187).  Under the terms of the evolving recommendation on the scope of the employment relationship, workers belonging to the latter two groupings would, at best, fall into the categories of ambiguous and disguised but workers in the first group are likely viewed to be engaged in “genuinely commercial arrangements.”  Still, regardless of the form of engagement, workers in each group may resemble paid employees.

Click on the picture to enlarge

While there are four dominant forms of self-employment in Canada (Figure 17a), full-time solo self-employment[39] is the best test-case for assessing ongoing international efforts to stretch the employment norm to accommodate dependent workers lacking labour protection on account of a lack of employee status.  Full-time solo self-employment constitutes half of total self-employment in Canada.  This is true among men and among women (Figures 17b and 17c), which is notable since self-employment has historically been the preserve of men, especially those forms exhibiting genuine qualities of entrepreneurship.[40] 

Click on the picture to enlarge

Click on the picture to enlarge

Click on the picture to enlarge

One of the chief objectives of the evolving international labour standard on the scope of the employment relationship, now under discussion, is to limit contractual manipulation designed to “deny certain rights and benefits to dependent workers” and another is not to interfere with “genuine commercial and independent contracting arrangements,” in other words, arrangements that reflect entrepreneurship (ILO 2004, par 5).  Negotiations are proceeding on the basis of a tenuous balance characterized by employers' refusal to extend the scope of labour protection and workers' collective desire to ensure that self-employed workers – as distinct from entrepreneurs (Fudge, Tucker and Vosko 2002) – receive the protection that they are due.  The compromise reached thus far attempts to balance these competing aims by devising guidelines for eliminating disguised employment or employment relationships that are falsely given the appearance of a different legal nature normally commercial but also often civil, cooperative, or family-related.  Statistically, reasons for pursuing self-employment are the best measure available for discerning the degree to which full-time solo self-employment reflects entrepreneurial values.  The self-employed are often depicted as choosing independence, freedom, and autonomy over security (Lin, Yates and Picot 1999, 6).  However, many pursue self-employment for other reasons.

While 44 percent of men and 28 percent of women in this form of self-employment cite “independence and freedom” as their chief reasons for self-employment, a greater percentage – 56 percent of men and 72 percent of women – cite other reasons.  Despite the widespread assumption that the self-employed are driven by entrepreneurial values, almost equal percentages of men (26 percent) and women (23 percent) pursue full-time solo self-employment because they cannot find suitable paid work.  For others, care-giving responsibility is another common reason.  Among women in full-time solo self-employment, 32 percent cite “balancing work and family” and “work from home” as their main reason for self-employment – five times the number of men (Figures 18a and 18b).  More women pursue full-time solo self-employment for care-giving reasons while more men pursue it for entrepreneurial reasons.  These patterns highlight the importance of examining the extent to which full-time solo self-employment not only resembles paid employment, but is precarious. They also underscore the importance of probing how various aspects of precariousness experienced by men and women in this group relate to the gender contract.

Click on the picture to enlarge

Click on the picture to enlarge

Access to a package of social benefits and entitlements akin to the social wage is one feature of high quality of employment.  Among the self-employed, a good indicator is extended benefits coverage.  In general, those in full-time solo self-employment have low levels of benefit coverage and men and women lack benefits in equal measure; in 2000, 43 percent of men and 45 percent of women had no benefits whatsoever and 61 percent of full-time solo self-employed people lacked extended health coverage, a rate that also holds equally for both sexes (Figure 19). 

Click on the picture to enlarge

Exploring the source of extended health coverage clarifies this picture further.  Among those with extended health coverage, spousal coverage is quite common.  As a source of coverage, it is considerably higher for the full-time solo self-employed (57 percent) than those engaged in full-time employer self-employment (20 percent).  Yet there are significant gender differences among the full-time solo self-employed: 70 percent of women in contrast to 49 percent of men access these benefits through a spouse (Figure 20).  Not only do many more women than men in the full-time solo group pursue self-employment because of care-giving responsibilities, many more women than men confront dependency as a consequence. 

Click on the picture to enlarge

The absence of benefits coverage generally, and independent access to benefits more specifically, leads to insecurity for many full-time solo self-employed people, especially women.  But an even truer test of whether a self-employed person is a genuine entrepreneur is income level.[41] Beyond employer status, among the full-time solo self-employed, income from self-employment is key to identifying those in legal arrangements differing from an employment relationship but resembling paid employees (Fudge, Tucker and Vosko 2002, 99).  Income level[42] is also a good indicator of precarious employment amongst the self-employed, especially when analyzed in relation to benefits coverage and source of benefits coverage.  Fifty-six percent of full-time solo self-employed people have annual incomes of $40,000 or less in comparison to 64 percent of full-time permanent employees.  And 25 percent earn less than $20,000 in contrast to 16 percent of full-time permanent employees.[43] Gender differences are particularly stark amongst those in the lowest income group: the percentage of women falling into the lowest income group is double that of men.  In both income groups, the full-time solo self-employed resemble paid employees.  They do not have paid help and they have income levels similar to (or lower than) paid employees. 

Click on the picture to enlarge

Combining benefits coverage and income level, the percentage of male full-time self-employed employers with incomes of $40,000 or less and no benefits (15 percent) is around half the size of the percentage of male full-time solo self-employed (29 percent). Amongst the full-time solo self-employed, the percentage of men and women with incomes of $40,000 or less and no benefits is equal.  Yet considering benefits coverage in isolation obscures gendered dependency among the full-time solo self-employed.  Sharp differences emerge in comparing men and women earning incomes of less than $40,000 and lacking independent access to benefits.  Full-time solo women are much more poorly off than men: among these women, 57 percent have incomes of $40,000 or less and have no benefits and/or no independent access to dental and health benefits[44] in contrast to 45 percent of men.  As Rooney et al (2002, 5) argue, “although self-employment allows women to work more autonomously and may provide the flexibility needed to accommodate family-related obligations, the lower incomes associated with self-employment, and the instability in income associated with fluctuations in the demand for products and services places many self-employed [women] in a precarious financial situation.” Among those in the full-time solo group most resembling employees, precariousness is highly gendered.

Regardless of whether contractual manipulation is the source of the problem (i.e., the root of exclusion from labour and social protection), a large percentage of full-time solo self-employment resembles paid employment, a sizeable segment of which is highly precarious.  In addition to lacking social benefits and statutory entitlements, from maternity leave[45] to, in many instances, employment insurance coverage, and normally collective bargaining rights, half of full-time solo self-employed earn less than $40,000 and have no benefits and/or no independent access to benefits.  Even more striking, 30 percent earn less than $20,000 and have no benefits and/or no independent access to benefits (Figure 23a) – and among women the percentage is 45 (Figure 23b). 

Click on the picture to enlarge

Click on the picture to enlarge

Click on the picture to enlarge

Click on the picture to enlarge

The sharp differences between men and women in this highly precarious group of workers reflects the continued dominance of a male breadwinner/female caregiver gender contract – many more women than men pursue self-employment on account of care-giving responsibilities.

1. The Status-Modified Employment Norm in Context

Workers lacking labour protection because their work arrangements are given the appearance of a different legal nature than an employment relationship in order to deny them certain rights and benefits are the main target of ongoing efforts at international labour regulation.  The long-term goal of standard-setting on the scope of the employment relationship is to improve the situation of people cast as self-employed that are actually dependent workers (i.e., they resemble paid employees).  Eliminating disguised employment is the starting point.  And the premise that “the conditions governing the method, time and place of the performance of services may not bear any similarity to the elements considered by the courts of a relationship of this kind” underlies this goal; so do concerns about the precarious situation of dependent workers, especially women, on account of recent developments in employment relationships (ILO 2000b, par 14; see also ILO 2003a, 53).  Thus far, the approach to re-regulation under consideration entails devising international guidelines for the development of national policies fostering access to labour protection among workers with an identifiable employment relationship.

Canada supported this approach at the International Labour Conference in 2003; it also advocated initiating the process of standard-setting with a recommendation on disguised employment.  Speaking on behalf of New Zealand, South Africa, and the United States, in a rare acknowledgement of problems prevalent in liberal industrial countries, the Canadian government representative successfully proposed the following wording for conclusions on the scope of the employment relationship:

One of the consequences associated with changes in the structure of the labour market, the organization of work, and the deficient application of law is the growing phenomenon of workers who are in fact employees but find themselves without the protection of an employment relationship.  This form of false self-employment is more common in less formalized economies.  However, many countries with well-structured labour markets also experience an increase in this phenomenon (ILO 2003b, par 32).

And yet the Canadian case highlights the limits of approaches addressing status-based deviation from the employment norm by narrowly targeting disguised employment.  Among the full-time solo self-employed, a sizeable percentage resemble paid employees, a significant subset of whom are precarious, especially women, whether or not their employment relationships are disguised.  So-called disguised employees who work for just one company, and whose status is a device to reduce total taxes paid by the firms and the workers involved, are virtually impossible to distinguish, not only statistically but also conceptually, from workers in “objectively ambiguous” situations (ILO 2003a, 2).  Furthermore, workers in both types of situations often have characteristics similar to self-employed contractors with multiple clients (i.e., income level, benefits coverage, source of benefits).  In Canada, so-called dependent workers in situations needing protection represent a much broader group of the self-employed than those in disguised employment relationships.  Recall that solo self-employment, itself a blurred category,[46] covers self-employed contractors who normally have multiple clients and may contract for a service or product; this group includes artists and cultural workers of various sorts (MacPherson 1999, Vosko forthcoming), fishers (Clement 1986), construction workers (MacDonald 1998), and truck drivers (Madar 1999). People labelled independent contractors that work directly for a contractor, such as newspaper carriers (Tucker forthcoming) also fall under this group.  So, too, do employees, such as rural mail couriers (Fudge forthcoming), who are explicitly excluded from the personal scope of labour protection, and collective bargaining, and are thereby disguised.  The latter group, however, only represents a limited segment of the solo self-employed.  The absence of an identifiable employment relationship is only one among several variables key to establishing whether a group is in need of labour protection.    

Guidelines for ensuring that persons within an employment relationship have access to the protection that they are due are the expected outcome of talks in 2006.  Evaluated against the Canadian case, international intervention of this sort could contribute to extending labour protections to some such workers, but it would do little for the many self-employed contractors in occupations or industries defined by these norms and practices.  Even if the evolving guidelines are over-inclusive, these workers will never fulfil criteria required to establish an employment relationship because of the nature of their profession or their trade. 

The Canadian example thus highlights the problems with taking the distinction between employees and the self-employed as a basis for labour protection.  In this context, many people in full-time solo self-employment are in need of the same labour and social protections enjoyed by paid employees, not simply those employees whose relationships are technically disguised.  As Judy Fudge, Eric Tucker and I argue (2002, 119-121), there is no principled reason for excluding any persons that are dependent on the sale of their capacity to work from the scope of labour law unless there is a compelling reason.[47]  The challenge is to design and adopt appropriate systems of extending protections and entitlements – ranging from access to collective bargaining rights, to social wage benefits and anti-discrimination procedures – capable of protecting this group.

Even if the goal is more modest – that is, limiting precarious work by addressing status-based deviations from the employment norm – the preceding portrait illustrates that policies aimed at combating disguised employment and clarifying ambiguous situations are insufficient.  Canada is nevertheless already following this path. In the last quarter of the 20^th century, the Canadian policy emphasis centred on refocusing the employment relationship.  The legal status of employment remains the entry point for most employment-related protection but policy-makers, courts, and administrative decision-makers have responded to the problem of the personal scope of labour protection in myriad ways (Fudge, Tucker and Vosko 2002, 8; for a detailed review see: England, Christie and Christie 1998, 2.1).  In some instances, legal tests that allow for an expansive category of employee have been adopted – this has often involved moving from tests resting on control and subordination to tests centring on economic dependence, a tendency growing in the post-1960 period (Arthurs 1965; Bendel 1982; Davidov 2002).  In others, they have minimized the significance of the distinction between employees and independent contractors by deeming persons not normally classified as such to be employees.  This has normally involved legislative or administrative action, illustrating that extending coverage to non-employees is ultimately a question of public policy (Fudge, Tucker and Vosko 2003b).[48] 

In the process, either by adapting tests or though a course of deeming, groups such as freelance journalists and homecare workers have gained access to collective bargaining rights (Cranford forthcoming; Vosko forthcoming).  Several Canadian jurisdictions have also minimized the salience of the employee/independent contractor distinction under human rights legislation (e.g., Federal, Nova Scotia and Prince Edward Island) and a majority have done so under occupational health and safety standards (Commission for Labor Cooperation 2003, 14; for a broader discussion of legislation focussing on social justice, see Fudge, Tucker and Vosko 2002, 65-73).  Furthermore, social policies, like employment insurance, have been extended with slight modifications to specific groups of workers, such as fishers (Schrank 1998, Vosko 2003),[49] as well as barbers, hairdressers and manicurists that are not employees and some taxi-drivers and drivers of other passenger-carrying vehicles (Fudge, Tucker and Vosko 2002, 81-82).[50]  The Canada Pension Plan also covers independent contractors but they are required to pay both employer and employee contributions. 

Under federal legislation on the Status of the Artist (1992), and parallel legislation in Quebec, collective bargaining legislation also extends to independent contractors that are professional artists.  This unique federal and provincial legislation represents the crest of innovation in Canada (MacPherson 1999; Vosko forthcoming). The more dominant approach has been to maintain the employee/independent contractor distinction while simultaneously extending coverage to workers resembling paid employees, who would otherwise fall out of this definition, through other means.

The boldest efforts to address forms of self-employment resembling paid employment in Canada cater to specific groups of contractors, such as artists, fishers and owner-drivers, while the remainder are concerned effectively with limiting what, in ILO and OECD parlance, are described as disguised and objectively ambiguous situations.  There has been virtually no attention to addressing the normative issue – that is, whether labour protection should extend primarily to employees (Fudge, Tucker and Vosko 2002, 95, 105-108).  This is the central weakness of the Canadian approach – a weakness underscored by the extent of precarious full-time solo self-employment.  However, there are other major shortcomings that relate to gender inequality. 

Indeed, the Canadian case illustrates the consequences of adopting an approach centred on minimizing status-based deviation from the standard employment relationship that retains male norms not only of employee status but entrepreneurship as well.  A sizeable proportion of full-time solo self-employment is precarious, and women in this category experience dimensions of precariousness disproportionately.  Granted, there is some recognition that many more women than men in self-employment endure financial hardship in Canada.  In 2004, the Prime Minister's Task Force on Women Entrepreneurs proposed that the federal government extend maternity leave benefits to self-employed women.  It noted that “women entrepreneurs would gladly pay into Employment Insurance if it meant that they would have access to these benefits” (Canada 2003, Recommendation 4.01, italics added).  And, it also observed that “many [women] are in lower income categories than their male counterparts,” and that their socio-economic situation unfairly compromises their ability to save for retirement (Canada 2003, Recommendation 12.06).[51] But despite these acknowledgments, the main thrust at the policy level is the promotion of women's entrepreneurship rather than extending labour protection to self-employed workers.

III.   Erosion of the Standard Employment Relationship

Precarious work and deviation from the standard employment relationship are often conflated.  The result is that efforts to limit precarious work frequently focus on forms of employment that in some way differ from the employment norm or its closest proxy – full-time permanent employment.  But is full-time permanent employment necessarily characterized by the security and durability associated with the standard employment relationship?  If full-time permanent employment does indeed reflect this employment norm in practice, then an approach to curbing precarious work centred on minimizing deviations from it offers promise.  If not, efforts to resuscitate the standard employment relationship, by crafting shorter-hours and place-modified employment norms as well as re-centering the employment relationship, are likely to be deficient.  The logical next step in evaluating the ILC model thus entails probing what is otherwise taken for granted as unproblematic – or, at least, less problematic.  It involves examining the quality and character of the standard employment relationship itself with attention to the shifting gender contract. The United States offers a suitable case study for this endeavour.

A. Full-Time Permanent Employment in the United States

Labour force trends in the United States indicate the resilience of time- and status-based deviations from the standard employment relationship, but the prevalence of part-time permanent employment and the size of full-time solo self-employment, and solo self-employment as a whole, are overshadowed by the deteriorating quality of full-time permanent employment. 

Recall that in the United States, part-time employment represents 19 percent of total employment (Table 1).  In this context, there is a shorter-hours employment norm, and it is highly gendered.  Fully 14 percent of total employment is part-time permanent employment and 20 percent of all employed women, in contrast to just 8 percent of men, hold part-time permanent jobs.  Furthermore, 38 percent of women working fewer than 40 hours per week (i.e., fewer than full-time weekly hours) cite childcare or the care of other family members as their main reasons for working “non-standard hours” (USDOL 1999, 33).  Standing at 5 percent of total employment, full-time solo self-employment is also quite significant, including among women, whose share of total self-employment increased between 1976-2002 (USDOL 2004, Table 32).  At the same time, the United States has a much larger share of full-time permanent employment than Australia and Canada.  Full-time permanent employment constituted 72 percent of total employment in the United States in 2001 in contrast to 47 percent in Australia and 63 percent in Canada and a greater percentage of women hold these jobs than in the past (Figure 17a; see also USDOL 1999, 35-36).  Indicative of this trend, the proportion of men and women working more than 40 hours per week on average has grown: according to the U.S. Department of Labour (1999, 36), between 1969 and 1998, it rose from 35 to 40 percent for men and from 14 to 22 percent for women.  Furthermore, one-quarter of men and one-tenth of women worked more than 50 hours per week (Jacobs and Gerson 1998, 458).

Click on the picture to enlarge

Trends in the quality and conditions of employment, such as the movement towards longer work weeks, affect how analysts characterize changing employment relationships, and thereby the descriptive concepts associated with the phenomenon of precarious work.  Politico-legal institutions and traditions, in turn, shape the meaning and substance of descriptive categories associated with employment – terms such as full-time and part-time, temporary and permanent.  In the 1980s, U.S. analysts coined the term “contingent work” as a shorthand moniker aimed at capturing the simultaneous growth of forms of employment and work arrangements differing from full-time permanent employment, such as part-time permanent employment, full-time solo self-employment, and market-mediated work arrangements, and the declining security attached to paid employment more generally.  Shortly thereafter, the Fact-Finding Report of the Commission on the Future of Worker-Management Relations (1994a, 93) defined contingent work as “marginal job relations” or “job opportunities that diverge from full-time continuing positions with a single employer.”[52]  In practice, contingent work normally connotes temporary or transitory employment; the distinction between contingent and non-contingent pivots on permanency (Vosko, Zukewich and Cranford 2003, 17; Vosko 2000, Chapter 1).[53] The U.S. Bureau of Labour Statistics classifies contingent workers as people who expect their job to end (some with a pre-determined end-date), many of whom are employees, and the rest of whom are self-employed or independent contractors.[54]  A lack of certainty of ongoing work is the foremost feature of contingent work.  However, countless studies also demonstrate that contingent work is also frequently characterized by low income and a lack of control over the labour process – both key features of precarious work.[55] 

Legal definitions, customs, and conventions help explain the durability of the concept contingent work – and its emphasis on a lack of permanency.  They also provide important clues as to the nature of precarious work in the United States, specifically, the dynamics of an eroding employment norm.  In the face of high levels of full-time permanent employment, the focus on contingency reflects the virtual absence of legislation on unjust dismissal at the national and state levels.  Unlike most other industrialized democracies and many industrializing countries, the United States has never had either a broad protection against unfair dismissal or discharge without just cause or any period of notice through the common law or by statute.  Rather, employment at will prevails – an employee can be discharged legally at any time “without notice for good reason, bad reason or no reason” (Commission for Labor Cooperation 2003, 26).[56]  The central consequence, put succinctly by Summers (2000, 69), is that while permanent employment is indefinite employment, “indefinite employment [is], by definition employment at will.”[57]

The significance of employment at will has varied over time.  Its legal meaning, and thereby its effects on the security and durability of full-time permanent employment, has progressed through several phases in parallel with the rise and decline of a particular version of the standard employment relationship, fostered by the growth of internal labour markets (Edwards, Gordon and Reich 1982; Doeringer and Piore 1971) and large vertically-integrated firms (Hyde 1998; Stone 2001), and a particular version of the male breadwinner/ female caregiver contract (Applebaum 2001; Fraser 1997).  After World War II, implicit contracts for lifetime employment[58] grew to dominance in the United States.  In this first phase, employment at will prevailed but employers “routinely entered into contracts in which people were effectively guaranteed lifetime employment” (Hyde 1998, 3).  Underpinning this practice was an implicit bargain between workers and employers that firms would invest in workers' acquisition of skills and knowledge, provide workers with a range of social benefits and entitlements, including back-loaded benefits such as pensions, and increase workers' wages incrementally – all in exchange for loyalty over the long term (Jacoby 1985; Stone 2001). 

The lifetime employment model was relatively short-lived in the United States, however.  It waned in the 1980s with the contraction of internal labour markets, the break-up of vertically integrated firms, falling real wages, and declining rates of unionization.  To cushion the potentially severe effects of employment at will, three judicially-created exceptions emerged: the public policy exception, the handbook rule, and an exception based on the covenant of good faith and fair dealing, “an implied obligation in every contract” (Summers 2000, 72; see also Hyde 1998; Sheehan 1997).[59]  And state courts gradually recognized these means of challenging employee discharges, thereby “transforming their traditional hands-off posture towards employment at will” (Dunlop Commission on the Future of Worker-Management Relations 1994a, 108).  However, after a brief hiatus in the 1980s, as Hyde points out, “sharply accelerating rates of job separation, increases in those responding to surveys that they were involuntarily terminated, and the new phenomenon of intentional decreases in the size of the workforce previously found only in troubled companies” coincided with this greater recognition of “a variety of causes of action by which discharged employees could challenge their discharges” (Hyde 1998, 101; see also Block and Roberts 2000, 293). 

Since the early 1990s, the positive impacts for employees of judicially-created exceptions on unfair dismissal or discharge without just cause have grown weaker.  Some judicial exceptions still hold sway but only one U.S. state – Montana – has a Wrongful Discharge from Employment Act (1987) that provides all non-union employees broad legal protection against wrongful dismissal.[60]  Legislatures are at an impasse where the adoption of new statutes is concerned and, with the decline of job stability (Swinnerton and Wial 1995), protection against unfair dismissal remains more limited in practice in the United States than in both Canada and Australia.[61]  While many American unions continue to negotiate collective agreements prohibiting dismissal without just-cause, enforcement is also a major problem.[62]

In the United States, and arguably elsewhere, full-time permanent employment does not necessarily provide the security, durability, and continuity associated with a standard employment relationship.  In a context of shifting employment norms where there is employment at will, continuity in full-time permanent employment is far from given.

In addition to the reasonable expectation of continuing employment, the standard employment relationship, as a norm, is characterized by social benefits and statutory entitlements that complete the social wage, union coverage, and a wage sufficient to support not only the worker but his or her dependents.  The high level of uncertainty surrounding permanent full-time employment in the United States should now be evident.  Yet what are the other characteristics of full-time permanent employment[63] in the contemporary United States and how are they gendered?

Medical benefits are among the most important social benefits historically linked to employment in the United States, in contradistinction to liberal industrialized countries like Canada, where core medical coverage is neither attached to the contract of employment, nor the employer, nor the workplace.[64]  Most Americans with health benefit coverage (i.e., medical care, prescription drug, dental, and vision coverage) access it through plans provided by their employers and requiring employee contributions.[65]  Public health insurance is available to the elderly through Medicare, which provides virtually free hospitalization insurance and low cost medical insurance covering physician services for people that are eligible for social security retirement or disability benefits.  It is also provided to a segment of people with low-incomes through Medicaid, which, according to the federal government mandate, must cover hospital visits, physician care, dental surgery, and other expenses, and may cover additional services depending upon the policy of a given state.  The remainder of the population must either secure benefits through an employer, self-insure (a practice common among the self-employed), or go without these benefits altogether.[66] 

While full-time permanent employees have significantly higher levels of independently-accessed health benefits coverage than their part-time counterparts (Jacobs and Gerson 2004; USDOL 1999),[67] their rates of medical coverage provided by an employer are declining.[68]  Between 1989/90 and 1998/99, their coverage rate declined from 83 to 68 percent (Figure 25).  In private industry, where coverage is highest in white-collar occupations and lowest in service occupations and where large establishments are more likely than smaller establishments to offer health insurance, an even higher percentage of full-time permanent employees lack medical care coverage (44 percent).  High percentages of full-time permanent employees in private industry also lack dental coverage (60 percent) and vision coverage (77 percent) (Figure 26).

Click on the picture to enlarge

Click on the picture to enlarge

The erosion of the standard employment relationship illustrates the consequences of delivering such fundamental social benefits as core health benefits primarily via the contract of employment.  Alongside chronically low levels of independent access to benefits among part-time employees – even among those in full-time permanent employment – access to basic health benefits, such as medical care coverage, became more limited in the post-1970 period.

Union coverage, another element of the standard employment relationship, is highly correlated with decent wages and social wage protections in the United States.  It has also long served as a primary means of securing protection against unfair dismissal or discharge without just cause.  Yet union membership declined dramatically in the post-1970 period, particularly in the 1980s and the 1990s, such that the proportion of private sector non-agricultural employees who are union members is less than one-third of that covered in the 1950s.  Among all wage and salary workers, union membership declined from 20 percent in 1983 to 13 percent in 2002.  The pace of decline was especially dramatic for men, whose membership rates dropped from 25 to 15 percent although women's membership rates still remain lower, standing at 12 percent in 2002.

Click on the picture to enlarge

Even among full-time permanent employees, rates of union membership are very low – only 14 percent have the protection of a collective bargaining agreement.  These trends are also gendered: 13 percent of women, in contrast to 15 percent of men, are unionized (Figure 28).

Click on the picture to enlarge

Low rates of union membership among full-time permanent employees in private industry in the United States are the result of several interrelated trends in worker representation and collective bargaining under the National Labour Relations Act (1935).  These trends include the small, and declining, number of National Labour Relations Board (NLRB) certification elections held and the small number of workers involved in successful certifications relative to the number of workplaces and employees.  As the Fact-Finding Report of the Commission on the Future of Worker-Management Relations (1994a, 67, emphasis added) observed:

The extent of NLRB election activity has trended downward through much of the post-World War II period.  In the early 1950s for example, the Board conducted nearly 6,000 elections, involving over 700,000 workers.  By the 1970s, the total number of certification elections had risen to over 7,500, but in smaller-sized units totalling 490,000 employees.  From 1975 to 1990 the number of elections fell by 55 percent to 3,628 elections involving 230,000 workers.

Fewer workers were involved in the NLRB representation process annually throughout the l990s and early 2000s than in preceding decades despite an expanding workforce; the small percentage of full-time permanent employees in private industry that are unionized reflects the declining number of workers unionized through NLRB elections.  As Kate Bronfenbrenner (2003, 32) has shown, “for the last two decades, unions have been able to gain representation for fewer than 100,000 workers each year, far fewer than the 400,000 union jobs that are lost each year from plant closings, layoffs, corporate restructuring, de-certifications, and contracting out.”  Furthermore, for those that manage to secure representation by a union, establishing collective bargaining, by way of a written agreement from the employer, is by no means secure; while analysts providing estimates for the 1950s report that 14 percent of efforts to secure a first contract failed over that decade, estimates of union failure grew in the 1980s and ranged from 20 to 37 percent (St. Antoine; and, Pavvy, as cited in Dunlop Commission 1994a, 73).  And, by 1994, about a third of workplaces that voted to be represented by a union did not obtain a collective bargaining contract with their employer (Dunlop Commission 1994a, 79).  In the last decade, according to Brofenbrenner (2003, 48) the situation has grown even worse such that “the overall private sector first contract rate is only 60 percent.”[69]

A large percentage of Americans are full-time permanent employees, many of whom, especially women, are working longer weekly hours than in the past.  Consistent with this trend, the percentage of married couples in which both spouses work more than 40 hours a week rose from 3.6 percent in 1969 to 10.1 percent in 1998; furthermore, almost 7 percent of couples with children under 6 were in this situation in 1998 in contrast to just 2 percent in 1969 (USDOL 1999, 35-36).  Falling and/or stagnating wages and increasing wage inequality help explain these trends.  While real hourly earnings rose by more than half for production and non-supervisory workers in private non-agricultural industries after World War II, when the male breadwinner/ female caregiver norm reached its height at over half of all married couples, most of this growth occurred in the 1950s and 1960s (Figure 29).  After peaking in 1973, real hourly earnings fell or stagnated for two decades – and they only began to stabilize in 1998.  Wage inequality also grew in the 1980s and 1990s; according to the United States Department of Labor (1999, 19), “after forty years of narrowing inequality, the high-to-low wage ratio[70] increased by 19 percent between 1979 and 1999 (from 3.7 to 4.4), largely because low-wage workers' earnings fell dramatically [in the 1980s].”

Click on the picture to enlarge

Among all full-time permanent employees, average weekly wages stood at $788.70 in April 2004.  Despite rising hours of work among this group, particularly among women, men earned $873.40 on average while women earned just $681.80 – a weekly wage gap of 78 percent (Figure 30).  There is a strong relationship between average weekly earnings of men and women, marital status, and presence or absence of young children.  Among full-time permanent employees, average weekly wages are lower than the national average for both married and single people with children under 6.  This is especially the case for women.  The average weekly wages of women in full-time permanent employment who are married with children under 6 are lower than the average for all women ($675.60) and they are lowest for single women with children under 6 ($424.30), who comprise the vast majority of single parents in the United States (Figure 31).

Click on the picture to enlarge

Click on the picture to enlarge

Many full-time permanent employees not only lack the protection afforded by a collective agreement, and the formal protections against unjust dismissal and social wage benefits flowing normally from such protection, but earn hourly wages that leave them in poverty.[71] An hourly wage of $8.70 or less is a conservative measure of low income in the United States as it amounts to $18,100 annually, the current official poverty level for a family of four (Schulman 2003).  Exploring low income and union status together, Figures 32a-c depict the percentage of full-time permanent hourly employees that are not unionized and earn less than $8.70/hr.  In April 2004, 20 percent of all full-time permanent hourly employees were in this situation.  One-fifth of all full-time permanent hourly employees – and fully a quarter of women –  experience precariousness along these two dimensions (Figures 32a, b and c). 

Click on the picture to enlarge

Click on the picture to enlarge

Click on the picture to enlarge

1.       Erosion of the Standard Employment Relationship in Context

In the United States, the rise of precarious work takes sharp expression in the deterioration of the normative model of employment that reached its apogee in the era of the lifetime employment model.  Employment that is full-time and permanent still dominates.  As the preceding section illustrated, however, many full-time permanent employees lack certainty of continuing work, social benefits, such as employer-provided health care coverage, and protections afforded by a collective agreement.  Many also confront low income and women, especially single women with young children, are highly vulnerable to precariousness.  As Alan Hyde (2000, 8) observes, “it is entirely possible in the U.S. to be [a full-time indefinite] ‘employee,' and yet be employed at will, have no legal or factual expectation of continued employment, no union, no practical way of obtaining union representation, no health insurance or pension.”  In the face of declining and/or stagnating real wages and job instability, these trends are contributing to the bifurcation of the labour force and longer weekly hours.  Considering total employment, most American men work longer weekly hours than men in most other countries.  The same is true for women – and the percentage of women working over 50 hours per week is higher in the United States than any other country in the world (Jacobs and Gerson 1998, 458-59).  Although many women still work part-time, rising weekly hours among women[72] and the high prevalence of dual-earner households are wholly consistent with the trend towards longer total hours.  The case of the United States highlights the complex character of precarious work in liberal industrialized countries: not only is the standard employment relationship declining numerically, its quality is deteriorating and thus its social meaning is changing. 

The ILO model provides limited guidance for countries, like the United States, where many full-time employees are enduring longer weekly hours as well as a low degree of regulatory and social protection and low income. Because their strategy is to stretch the standard employment relationship, new and emerging international labour standards on part-time work, home work, private employment agencies, and the scope of the employment relationship are ill-equipped to empower national governments to limit erosion of the form of employment that falls within the traditional standard.  The Social Declaration offers some potential as a tool for change as it endorses the Convention on Discrimination and the Convention on Freedom of Association and the Right to Organize.  However, as the American example so vividly illustrates, the equal treatment paradigm underpinning anti-discrimination legislation – in this case nationally but also in the ILC – is not designed to address an eroding employment norm.  Stronger minimum standards are required, such as those on hours of work and wages, but the Social Declaration – like anti-discrimination legislation in the United States – is limited by its procedural orientation.  The Social Declaration also promotes freedom of association and the right to organize and bargain collectively.  In this instance, the American case highlights the limits of a promotional standard that assumes a correlation between regulatory effectiveness and the presence of labour rights – technically, most full-time permanent employees have the right to freedom of association and the right to organize and bargain effectively in the United States, although they face significant hurdles in securing union certification, first contracts, and/or the enforcement of collective agreements.  Not surprisingly, then, in its Fact-Finding Report, the Dunlop Commission (1994a, 23) found that while “the number of statutes affecting the workplace…have increased significantly over the past twenty-five years…the appropriations for organization and staff to secure enforcement have not kept pace with the enlarged responsibilities of federal agencies.”

In other areas, some initiatives for change that address the erosion of the employment norm and growing instability in the gender contract are emerging.  Among the most promising are initiatives related to working time, broadly conceived.  These initiatives take three distinct forms.  Two respond to the erosion of the employment norm and to what is increasingly described as growing work-family conflict: “work-facilitating reforms” (Gerson and Jacobs 2003, 464-465) and “family-friendly” reforms (Rosenfeld 2001, 105).  The goal of work-facilitating reforms is to limit barriers to employment, especially for dual-earning parents and single mothers; employer-sponsored daycare, after-school programs, and syncing the workday and the school day are examples.  “Family-friendly” reforms, in turn, aim to encourage employers to support workers in spending more time with their families in the private sphere, especially their children.  They range from flex-time, job sharing, and provision for working at home to unpaid leaves for care and they are, by far, the most prominent type of work-time reforms in the United States (Applebaum 2002a; Rosenfeld 2001). 

Introduced in 1994, the Family and Medical Leave Act epitomizes reforms of the “family-friendly” variety.  This legislation requires federal public employers and private employers that maintain fifty or more employees on their payroll over 20 calendar workweeks in a given year to allow employees to take up to twelve weeks of unpaid family leave for medical reasons, the birth or adoption of a child, or for the care of a child, spouse or parent who has a serious health condition.[73]  There are several important limitations to the Act.  Many part-time and temporary employees, and most independent contractors, cannot access protection because employees are only eligible for this leave if they have been employed by the employer for twelve months and have at least 1,250 hours of service.  Many parents and people with dependents, especially women, are employed in firms with fewer than fifty employees.  Like Australia, the leave is unpaid and, as a consequence, since women on average earn less than men, they are more apt to take leave, exacerbating “a cumulative cycle of low wages and increased career interruptions that [in practice] inhibits job mobility for women” (Jacobs and Gerson 1998, 466).  Finally, while some family-friendly policies are cast as enhancing ‘choices' for employees, their individualistic orientation neglects fundamental structural issues – they do little to alter the culture of long work hours and leave rigid job structures intact.  There is still ‘no such thing as a part-time career.'

More compelling than either work-facilitating or family-friendly policies are working time reform proposals that relate to the length of work hours, and thereby to job structures themselves.  In the United States, no such reforms have been introduced but proposals that speak directly to structural issues, such as addressing long weekly hours and the bifurcation of hours, are mounting.  Jerry Jacobs and Kathleen Gerson (1998, 466) make three promising proposals[74] premised on their argument that the current structure of working hours “forces some people to either accept longer hours than they would prefer or pay a substantial penalty in terms of career mobility.  It compels others to accept part-time jobs with fewer hours than they would prefer and lower benefits, job security, and economic resources than they need.”  Their first proposal is to extend the Fair Labour Standards Act (1938) to employees that are professionals and managers.  At present, these groups do not fall under this Act and are therefore not subject to existing wage- and hours-legislation.  In practice, this means that employers are not required to pay for overtime.  Since professional and managerial employees are at high risk of excessive weekly hours, they argue that extending this Act would reduce overwork among salaried workers.  Their second proposal is to shorten the standard work week slightly – to 35 hours – for all employees by providing tax incentives for employers to adopt shorter hour weeks, such as reduced payments into mandatory employer contribution systems.  Their third, and final, proposal involves the introduction of mandatory prorated benefits for all employees – and while it focuses on the American context, this proposal could have application beyond it.  Its goal is twofold: to discourage excessively long hours among salaried employees and to reduce working-hours polarization.  The idea is to provide all employees with mandatory benefits[75] that accrue on the basis of hours of work.  As it stands, full-time workers do not receive additional increments in benefits if they work more than forty-hours per week.  At the same time, even if they work close to a full-time or standard work week, part-time employees often do not receive any benefits.  If this type of proposal were to be adopted, it could reduce the tendency among employers to require salaried employees to work excessive overtime because there would be no cost savings in the provision of benefits. It would also improve the situation of part-time employees, many of whom work weekly hours equivalent to a standard full-time work week elsewhere. Granted, this proposal for prorated benefits has several potential pitfalls, specifically those identified in the discussion of the limits of a shorter-hours employment norm extending pro-rated benefits to part-time employees without altering fundamentally the norm itself.  The question “by whose standard?” (Fudge and Vosko 2001b) remains critical but, in the American context, using prorated benefits to limit excessive hours of work could curtail growing economic inequality and improve gender equality.

The latter two proposals resemble alternative approaches to limiting precarious work emerging elsewhere, principally those in Europe at the national and supranational levels – initiatives and alternative visions that merit greater scrutiny.

20. The data sources for the figures and tables in this section are as follows: Figure 1 -- OECD Labour Force Statistics:  Indicators; Table 1; Figure 2a, Figure 2b, Figure 2c, Figure 3, Figure 4, Figure 15, Figure 24 – Australia – 2000 Survey of Employment Arrangements and Superannuation, Special Request, Australian Bureau of Statistics; Canada – 2000 Labour Force Survey, Public Use Microdata File Custom Tabulation, Statistics Canada; United States – February 2000 Current Population Survey and Contingent Worker Supplement, Public Use File Custom Tabulation, United States Bureau of Labor Statistics; Figure 5-14 -  Australia – 2000 Survey of Employment Arrangements and Superannuation, Special Request, Australian Bureau of Statistics; Figure 16-23 – Canada 2000 Survey of Self-employed, Public Use Microdata File Custom Tabulation, Statistics Canada; Figure 25 -26 – William J. Wiatrowski, May 26, 2004, “Documenting Benefits Coverage for all Workers”, Compensation and Working Conditions Online, USDOL; Figure 27 – “Women in the Labor Force: A Databook” USDOL, Feb 2004 p. 70-71; Figure 28, Figure 30-32 – April 2001 Current Population Survey, Public Use File Custom Tabulation, United States Bureau of Labor Statistics; Figure 29 – futurework: Trends and Challenges for Work in the 21st Century, 1999, USDOL.

Data referring to 2000 for the United States refer to data retrieved from the Current Population Survey and the Contingent Work Supplement of February 2001.

21. This approach was first developed by the author, along with Zukewich and Cranford, as a critique of traditional approaches to conceptualizing and measuring ‘non-standard work’ and ‘contingent work’ in Canada (Vosko, Zukewich and Cranford 2003).  The discussion that follows applies this approach to Australia and the United States and develops it further (see especially Table 1).

22. In Canada, temporary work includes seasonal, contract, casual and ‘other temporary’ work, including work through a temporary agency.  In the United States, it includes all employees that describe their employment situation as temporary.

23. Casual employment is defined by the Australia Bureau of Statistics (ABS) as all paid employment lacking leave entitlements (both sick leave and paid annual leave).  Following the justification used originally by the ABS, I use the lack of leave entitlements to define casual employment to reflect the long tradition in Australia’s industrial relations system where casual employees are to receive “loadings” or precarity pay in lieu of entitlements for paid holiday leave and sick leave under federal awards (ABS 2001, par 4.38; see also Tham 2003, 3).

There is a growing movement to redefine the category casual in ABS surveys to conform to what O’Donnell (2004) labels its “vernacular” meaning – that is, intermittent, irregular, or on-call.  This tendency is evident in recent surveys that permit analysts to adopt the concept “self-identify casual,” such as the Forms of Employment Survey and the Survey of Employment Arrangements and Superannuation, to capture the so-called ‘true casual.’  However, this approach is questionable because the vernacular conception of ‘true casual’ has very little meaning at either a contractual or regulatory level.

The significance of the category casual and its legal meaning are discussed at length below.

24. See also Campbell (1998, 108), who highlights the similarities between casual employment in Australia and temporary employment in the EU.

25. This grouping is also consistent with that used by the OECD in defining temporary employment in Australia for the purpose of comparison. 

26. As Belinda Probert (1997, 186) illustrates, in 1933, 25.2 percent of women were in full-time employment and, in 1994, the figure was just 27.1 percent. 

27. The category fixed-term is used to refer to employment on either a fixed-term basis or paid by an agency since the latter is such a small group.

28. The statistical category part-time ongoing casual includes all part-time employees who lack leave entitlements (i.e., are not entitled to paid holiday or sick leave) that are neither on a fixed-term contract or paid by an agency.  Among this group, 74 percent expect to be with their current employer in 12 months time.  This definition is inspired by the findings of other earlier studies which reveal the long job tenure of many casuals in Australia (see for example, Campbell 1998, 72).

29. Except where noted otherwise, the ensuing discussion of the status of casual employees at the regulatory and contractual levels refers to federal examples. 

30. See also: Award Simplification decision (Full Bench AIRC, 23 December 1997, P7500), which uses the language “a casual employee is an employee engaged as such.”

31. At the state level, the movement to grant casuals family leave has also been incremental.  For example, the Queensland Industrial Relations Act (s. 16, 39 (2), 40) extends family leave, caregiver’s leave, and bereavement leave to ongoing casuals, defined as “a casual employee engaged by a particularly employer on a regular and systematic basis, for several periods of employment during the period of at least a year” (S. 15 (A)).  Yet most other states still exclude casuals from access to parental leave (e.g., South Australia and Western Australia).

32. The South Australian Clerks Award and the Metal, Engineering and Associated Industries Award are cases point.  For incisive commentaries on these awards, see Owens 2001 and Poccock et al 2004.

33. See: Automotive, Food, Metals, Engineering, Printing and Kindred Industries Union – re application for variation of award - casual employees - T4991 - 29 December 2000, AIRC.

34. This section draws from, and builds on, research conducted with Judy Fudge and Eric Tucker on the legal concept of employment and with Nancy Zukewich on gender and precarious self-employment (see Fudge, Tucker and Vosko 2002, 2003a, 2003b; Vosko and Zukewich, forthcoming).

35. Statistically, self-employment is normally distinguished from paid employment by the mode of remuneration (i.e., the self-employed receive profits or income from self-employment and employees earn wages) (Elias 2000; Loufti 1991).  This distinction is the point of departure for this discussion.

36. Operating an incorporated business is also a good indicator of entrepreneurship, although it is not taken up here for reasons of sample size (For greater detail on this indicator, see: Fudge, Tucker and Vosko 2002 and Vosko and Zukewich forthcoming).

37. 40 percent of the solo self-employed reported receiving support from their clients in 2000 (Delage 2002, Table B.7).

38. According to a study conducted by the Canadian Policy Research Networks, 51 percent of the solo self-employed had five or fewer clients in 2000 (Lowe and Schellenberg 2001, Table 4.2).

39. The full-time group is selected as the case study for consideration as it is an indirect way to control for hours.

40. In Canada, the four forms of self-employment fall on a gendered continuum of precariousness.  This continuum moves from full-time employer self-employment to full-time solo self-employment to part-time employer self-employment to part-time solo self-employment.  Along several dimensions of precarious employment, full-time employers are least precarious, while part-time solo self-employed people are the most precarious on the majority of indicators. This continuum is gendered since men are concentrated in the full-time employer category and women are concentrated in the part-time solo category.

41. Data on income level refer to income from self-employment. Income is a good indicator of the economic situation among the self-employed since they derive a range of benefits from their employment status invisible in earnings data (Fudge, Tucker and Vosko 2002, 26).  While income may be underreported amongst the self-employed, these data still highlight broad patterns.

42. The Survey of Self-Employment allows for analysis of income by four major income groups: less than $20,000; $20,000-$40,000; $40,000-$60,000; and $60,000 plus.  The ensuing analysis considers the full-time solo self-employed with incomes under $40,000.  It also considers those with incomes of under $20,000 since this group is unequivocally precarious along the dimension of income level.

43. This figure is consistent with other studies.  For example, Lin, Yates, and Picot (1999) reveal that almost three quarters of self-employed women (72 percent) earned less than $20,000 in 1994 compared to 48 percent of women employees.

44. The Survey of Self-Employment 2000 does not include a question about source of disability insurance; hence, data on source of benefits refer exclusively to dental and health benefits.

45. As a consequence, the average maternity leave is much shorter for women in self-employment than paid employment.  According to a study by Katherine Marshall (1999), 80 percent of self-employed women returned to self-employment within the first four weeks after having a child, compared to only 16 percent of paid employees.

46. Data indicate that solo self-employment may, in practice, be larger than is often estimated since conventional measures fail to capture the degree to which people move between solo and employer forms.  The self-employed that do not hire others in a given reference year are normally classified as solo self-employed while the self-employed that hire others are considered employers.  Based on this definition, in 2000, 46% of the self-employed in Canada were employers. Yet, when the same group was asked whether they had paid help during a particular reference week, only 38% fell into this category. 

47. We include the following areas of labour regulation in our recommendations: anti-discrimination law, pay and employment equity legislation, occupational health and safety legislation, minimum standards legislation, collective bargaining legislation and social wage and social revenue legislation.  Yet we acknowledge that there may be relevant distinctions between different groups of workers, such as the nature of their relationship to the entity that purchases their service, and these distinctions should be taken into account in the design of instruments to provide labour protections to all workers regardless of type of income they receive.

48. The well-known example here is the extension of the personal scope of collective bargaining law to dependent contractors, that is, worker who are economically dependent (see Arthurs 1965; see also Bendel 1982; Davidov 2002).  

49. Fishers’ EI is financed by self-employed fishers and designated employers and qualifying requirements are organized on the basis of earnings rather than hours.

50. Among these groups, it is extended to those that are not employees and who neither own or operate the business nor own more than fifty percent of the vehicle and who are provided supports from the business operator.  In these cases, contributions are split between the owner or operator of the business and the self-employed person.

51. To remedy this problem, it called for addressing inequities in the RRSP system to improve women’s preparation for retirement.

52. In describing “‘contingent’ worker-management relations,’” the report defined the contingent work broadly noting that it “often includes part-time workers, some of whom are voluntarily part-time, some of whom would like full-time work, and some of whom are multiple job holders.  It also includes employees of temporary help agencies – who may be full-time workers – and some of the self-employed including “owner-operators” or independent  contractors with only a single contract of employment” (Dunlop 1994a 93).

.53. Contingent work is not synonymous with precarious work – indeed, it is much narrower.

54. Three definitions of contingent work are used in the United States, each pivoting on the degree of certainty of continuing employment.  The first definition includes all wage and salary workers who do not expect their job to last.  This corresponds with the Canadian definition of temporary work.  The second definition narrows the focus to employment of very limited duration by including only those wage and salary workers who expect to work in their current job for one year or less and who have worked for their current employer for less than one year.  The third definition broadens the second by including self-employed workers who expect to be, or have been, in their current employment situation for one year or less (Vosko, Zukewich and Cranford 2003, 17).  For detailed discussions of this mode of classification see also: Belman and Golden 2002; Polivka and Nardone 1989; and, Polivka 1996.

55. See for example, contributions to the following two edited volumes: Barker and Christensen 1998 and Carre et al 2000.

56. The employment at will doctrine dates to a treatise written by Horace Wood in 1877 seeking to distinguish American law from the English common law; according to Summers (2000, 67), this doctrine advances the premise that the “employer has sovereignty except to the extent it has expressly granted employees rights.”  

57. Where the gender contract is concerned, indirect consequences include that “workers in the US can be – and frequently are – fired if their family responsibilities interfere with their jobs” (Applebaum 2002b, 94).  Furthermore, employees do not have the right to refuse overtime, for example, even if they are asked to stay beyond their shift without sufficient notice to arrange for childcare or eldercare.

58. I borrow this term from Alan Hyde (1998), although similar concepts are used by numerous scholars such as Stone (2001) and Jacoby (1985).

59. The first exception involves allowing an employee defined to be at will to sue in tort rather than contact; this strategy emerged as an attempt to circumvent the doctrine entirely and it has been used widely to protect employees who are discharged for refusing to violate public policy or who are discharged when an employer refuses to pay benefits already earned or in response to, for example, the filing of claims for workplace injuries (Block and Roberts 2000, 293; Summers 2000, 70-71).  The second device involves using principles of contract to craft exceptions.  Its aim is to show that the employer has circumvented its own general rules of conduct, procedures, and practices, often found in handbooks or personnel manuals distributed to employees, in discharging a worker (Sheehan 1997, 324-325).  In such instances, an employee handbook outlining disciplinary procedures or assurances that employees will not be dismissed without just-cause are used as evidence.  This exception applies “the basic contractual principle that a person is bound by implicit promises in a course of conduct” (Summers 2000, 72); according to Hyde (1998, 101) it contributed to a growing tendency within the courts in the postwar period to “treat the employment contract as the total of all communications on the job.”  The third type of exception is also contractual, although its focus is demonstrating that the employer is bound by an implied contract of good faith and fair dealing common to all contracts (Hyde 1998; Sheehan 1997).  It has been used successfully in the discharge of long-service employees (Summers 2000, 73).

60. Despite an agreement in 1991 by the National Conference of Commissioners on Uniform State Laws on a Model Employment Termination Act, no other state has adopted similar legislation (Commission on Labor Cooperation 2003).

61. For comparisons of Canada and the United Sates, see Block and Roberts 2000; Commission on Labor Cooperation 2003.

62. Two pieces of federal legislation place some limits on this problem by cushioning the consequences of job loss among employees either displaced by plant closing or mass layoffs or discharged.  Enacted in 1989, the federal Worker Adjustment and Retraining Notification Act requires large employers (i.e., those employing 100 or more full-time employees or a 100 of more full-time and part-time employees who work an aggregate of at least 4,000 hours per week, exclusive of overtime) to provide 60 days advance notice of planned plant closings and mass layoffs.  And, the Consolidated Budget Reconciliation Act (1985) requires employers that provide a group health insurance plan (excluding churches, public employers, and employers that normally employ fewer than 20 employees) to offer participating employees and their spouses and dependent children the chance to continue coverage when they are discharged, except for “gross misconduct” – such coverage may continue for 18-36 months. 

63. Due to data limitations, the proxy used for full-time permanent employment in the ensuing analysis is full-time employment.  Aside from technical reasons, the rationale for adopting this proxy is that full-time temporary employment accounts for just 1 percent of total full-time employment; thus, even when full-time temporary and permanent employment are grouped together, a reasonably accurate picture of full-time permanent employment emerges.

64. In Canada, while there are various means of extending labour and social protections, medical care and health insurance flow from what Brian Langille (2002, 140) aptly labels a “citizenship platform,” which provides social infrastructure regardless of an individual’s labour force status.

65. For all wage and salary workers in private industry, employee contributions for family coverage averaged $228.98 per month for family coverage and $60.24 per month for single coverage in March 2003.  Average monthly contributions required of employees rose by 75 percent for both family and single coverage between 1992/1992 and 2002/2003 (USDOL 2003, 1).

66. Notably in 1997, 17 percent of non-elderly U.S. residents fell into this last category – that is, they did not have access to either public health insurance or employer-based plans.

67. In 2003, only 9 percent of part-time employees in private industry had independent access to medical care coverage, only 6 percent had dental coverage and only 5 percent had vision coverage (USDOL 2003).

68. Along with other commentators, Jacobs and Gerson (1998) contend that part-time permanent employment remains sizeable as a percentage of total employment because it provides an easy route for employers to avoid providing health coverage.  They argue further that fixed rates of health coverage for full-time employees encourage longer worker hours, especially among salary workers since paid overtime is not required.  In other words, the structure of employer-provided health benefits coverage shapes bifurcated trends in working hours as well as erosion in access to benefits.

69. Employees’ efforts to obtain representation are also increasingly thwarted by the threat of discharge or unfair discrimination.  Indeed, according to the Fact-Finding Report of the Commission on the Future of Worker-Management Relations (1994a, 79) “the probability that a worker will be discharged or otherwise unfairly discriminated against for exercising legal rights under the NLRA has increased over time.”

70. The wage ratio is measured by the Bureau of Labour Statistics as the ratio of a high wage worker’s earnings (in the 90^th percentile of the wage distribution) to that of the low wage worker’s earnings (in the 10^th percentile).  The figures cited here reflect weekly earnings ratios.

71. The ensuing discussion focuses on full-time permanent hourly employees for technical reasons. Yet in order to be able to generalize the analysis to all full-time permanent employees, I use indicators of a lack of control over the labour process and low-income level suitable for both salaried and hourly employees.  I also omit employee-sponsored health care benefits from the analysis since they are more common among salaried than hourly employees (Jacobs and Gerson 1998, 461).

72. Women’s average weekly hours stood at 36.1 in 1998.  In Australia, people with weekly hours above 35 are considered full-time and, in Canada, people with weekly hours above 30 hours are considered full-time.  However, in the United States, people whose weekly hours fall below 40 are deemed to be part-time.

73. Over the leave, the employer “must provide health care benefits at the same level and under the same conditions as if the employee were actively at work” (Commission on Labor Cooperation 2003, 26).

74. Jacobs and Gerson first set out these proposals in their article “Toward a Family-Friendly and Gender-Equitable Work Week” (1998, 466-469).  For further elaboration, see also Jacobs and Gerson (2004).

75. Jacobs and Gerson do not define mandatory benefits, although in discussing a standard package based on a forty-hour work week, they mention life insurance, health care benefits, and disability insurance.

<< Back to Table of Contents

PART FOUR: Limiting Precarious Work?  Alternative Approaches >