Home About Us Reports Research Paper 2002 The Legal Concept of Employment: Marginalizing Workers Page 10

About Us

Reports

Research Paper

The Legal Concept of Employment: Marginalizing Workers

Many legal scholars have concluded that the traditional concept of employee does not provide a satisfactory basis for delineating the personal scope of labour and employment law nor for determining eligibility for social wage programs (Benjamin 2002; Brooks 1988; Carlson 2001; Davidov 2002 a, b; Davies 1999; Davies and Freedland 2000a; Freedland 1995; Hyde 2000; Linder 1992; Stewart 2002; Supiot 2001). At the national level there have been a number of attempts to expand the scope of certain aspects of labour protection to economically dependent workers (Benjamin 2002; Davies and Freedland 2000a; Comparative Labour Law and Policy Journal 1999) and an expert report (called the Supiot Report after Alain Supiot the main author) issued by the European Commission has endorsed such an expansion throughout the European Union (European Commission 1999; Supiot 2001). The personal scope of employment and labour protection will be discussed at the ILO’s 2003 conference (ILO 2003), despite the 1998 defeat of the ILO’s proposed Convention on Contract Labour, which would have extended some elements of labour protection to dependent workers.[117] There is widespread agreement that the traditional legal categories - “employee”, “independent contractors”, “contract of service”, and “contract for services” - no longer fit with the economic and social reality of work relations. The challenge, according to Patricia Leighton (2000, 288), is to develop a legal approach which equates with work realities and which is rooted in robust empirical data, so as to ensure the genuine self-employed/entrepreneur is correctly, consistently and convincingly categorized.

The conventional solution to the problem of determining the personal scope of employment and labour legislation has been either to develop a new, more inclusive test to determine whether or not a worker is an employee or to adopt a new, more extensive definition (Davidov 2002 a, b; Stewart 2002). These techniques have been adopted in a piecemeal manner across Canada in a variety of specific legal regimes and the problems with them are identified at the end of Part Four. Another solution, which is gaining ground in Europe (Davies and Freedland 2000a; Supiot 2001), is to develop a new concept to determine the personal scope of employment and labour legislation. In Britain, the concept “worker”, which broadly corresponds to the civil law notion of “parasubordination” and the Canadian term “dependent contactor”, is an attempt to shift the boundary of labour protection to encompass those apparently self-employed workers who are not genuinely in business of their own account (Deakin 2002, 191). While this solution does not obviate the immediate need to draw distinctions between dependent workers and independent contractors, it is a step towards limiting the significance of the distinctions between different types of workers for the purpose of legal regulation. Commenting on the concept “worker”, Deakin (2002, 191) speculates that

the longer-term implication of its use may be to dissolve entirely the traditional boundary between employees and the self-employed, leaving only independent entrepreneurs (those with business assets and the opportunity to capture residual profits) outside employment law. However, what is of particular interest is that the response has not taken the form of a de-socialization of the employment relationship. On the contrary, it is an attempt to extend the logic of social protection to certain forms of self-employment.

The statistical portrait of the self-employed in Canada suggests the time already has come to consider dissolving the distinction between employees and the self-employed for the purposes of labour protection. Few of the self-employed resemble entrepreneurs since the overwhelming majority of them lack many, if not all, of the distinctive features of entrepreneurship – ownership, autonomy, or control over production. The majority of the self-employed much more closely resemble employees than they do entrepreneurs, although for legal purposes many would be classified as independent contractors and, as such, they would be denied the legal protection available to employees (see Part Two, Charts II.8 and II.9). But, as the data reveal, not all independent contractors are entrepreneurs and many independent contractors are dependent on selling their labour. The statistical portrait of the self-employed in Canada suggests that instead of trying to rehabilitate the legal distinction between independent contractors and employees as a basis for drawing the boundary between commercial and labour law, the distinction should be abolished. To the extent that there are relevant differences in the conditions under and arrangements by which employees and contractors sell their labour, these should be taken into account in the design of labour law, not its application.

Most legal scholars recognize that determining the scope of employment and labour regulation should begin with a consideration of the rationale for a particular regulatory regime and the range of work contracts to which it should apply (Carlson 2001; Davidov 2002 a, b; Davies and Freedland 2000a; Engblom 2001; Hyde 2000; Langille 2002; Linder 1992; Maltby and Yamada 1997; Perritt 1988). The process for determining the personal scope of regulation involves identifying the range of work relationships or contracts, establishing the normative goals of the regulatory regime, and assessing the variety of work contracts against the normative goals of different regimes. The first two steps involve developing typologies for classifying different work contracts and regulatory regimes. The third step is evaluative, providing normative and economic justifications for recommending a specific scope of application in a particular context. There are also some ancillary, more technical questions that relate to reducing uncertainty and the opportunity for shams and to how specific regulatory regimes have been institutionalized.

In response to the now commonplace observation that contracts for the performance of work take on a variety of forms, scholars have attempted to construct new typologies of work relations that move beyond the employee/self-employed dichotomy (Davidov 2002 a, b; Davies and Freedland 2000a; Englbom 2001; Supiot 2001). These typologies capture some of the principal dimensions, especially autonomy/subordination and economic independence/dependence, of these complex relationships. Given the range and diversity of work relations, it is inevitable that any typology, no matter how sophisticated, will involve a degree of arbitrariness. However, the purpose of a typology is to identify key features of work relations that correspond reasonably well with social reality and that relate to the purposes of employment and labour legislation.

One typology of work and employment relations that has gained a foothold in attempts to address the problem of delineating labour law’s domain is that of Paul Davies and Mark Freedland (2000a, 35; Engblom 2001). Emphasizing the traditional dimensions of subordination and economic dependence, they identified four categories of work contracts or relations: 1) employees traditionally conceived; 2) employee-like workers who perform work personally and are highly dependent economically on one or more employers; 3) people who perform work personally but who run an identifiable business of their own; and 4) people who have contracted to produce a result that may or may not involve personal work on their part. This typology is a useful starting point because it captures positions that are already identified to some extent in law and provides a reference point that allows for a comparison between different views about the appropriate extension of the personal scope of employment and labour law. However, it is important to recognize that these “conventional” legal categories are problematic because they represent arbitrary points on a spectrum that, as the prior discussion of legal tests demonstrated, adjudicators have found enormously difficult to apply in a consistent, predictable, and appropriate manner. Moreover, there are some cases in which an employee may be less economically dependent or subordinate than a so-called independent contractor (for example, an “in-house” company lawyer compared to a small contractor who runs a home renovation business). Thus, while the categories provide a useful heuristic, extreme caution is called for in making them legally salient for the purposes of determining the appropriate scope of labour law. Moreover, the question of category definition will have to be revisited once the normative, economic, and institutional justifications for delimiting the scope of labour law have been considered.

The typology presented below differs from Davies and Freedland’s typology in three respects. First, the terminology used to identify the different categories of work relations reflects terms that are commonly used in Canadian law and policy. The term dependent contractor covers what Davies and Freedland call employee-like workers; persons who perform work personally but are in business on their own account are referred to as independent contractors, and product sellers is used instead of the term “non-personal work”. Second, in order to indicate the relational nature of employment and to capture some of the diversity of employers, their fourth category, product sellers, has been subdivided into dependent and independent product sellers. Third, the categories in Davies and Freedland’s typology are contrasted with the traditional sociological categories that emphasize the ownership of the means of production for understanding the nature of power and authority relations between labour and capital. As discussed in Part One, proletarians are workers who do not own sufficient means to enable them to produce independently and so are dependent on the sale of their labour power for their survival. The bourgeoisie own sufficient means to employ proletarians, over whom they exercise control. In between is a group of petit bourgeois who are partially dependent on the sale of their labour power but also own some or all of the means of production and may hire a few workers. The absence of economic dependence on the performance of work (sale of labour) would lead to the categorization of an individual as a product seller, not an independent contractor. Most independent contractors, who would be classified as petit bourgeois in classical sociological terms, do not illustrate the distinctive traits of entrepreneurship – autonomy and ownership -- and thus are more akin to proletarians than they are to the bourgeoisie.

Click on image to enlarge

How does this typology measure up against the social reality of work relations? This assessment is complicated by the fact that the existing statistical definitions and data are based on the distinction between employment and self-employment and do not specifically address the distinctive features of entrepreneurship. However, despite this limitation it is still useful to estimate the size of the different categories since such estimates provide some indication of the number of people who will be affected by different extensions of labour law’s domain.[118] Drawing upon the data discussed in the portrait of the self-employed provided in Part One Figure I.1, in Canada 83.8 % of the work force are dependent workers (employees and dependent contractors) and 5.6 % are bourgeoisie or petit bourgeois (self-employed employers who include independent contractors as well as dependent and independent product sellers). The difficulty is classifying the 10.3 % of the workforce in Canada who are self-employed on their own account.[119] The crucial question is whether this group of self-employed are entrepreneurs.

Although there is no simple method for identifying the universe of entrepreneurs within the broader class of the self-employed in Canada, there are several measures that can be used to approximate entrepreneurship. These measures, which are employer status and income, provide some indication of the proportion of employed people who would be excluded if labour protection were extended to all independent contractors who are not entrepreneurs. If self-employed employers were considered to be entrepreneurs, according to Figure V.2-I, only 34.6% of the self-employed would be excluded from labour protection. This group amounts to 5.6% of the total number of people employed in Canada. If all the self-employed with annual incomes of over $90,000 a year were considered to be entrepreneurs and thus excluded from labour protection, only 10.0% of the self-employed would be excluded, which amounts to only 1.6% of all the people employed (see Figure V.2-II). As the measure of entrepreneurial status is made more rigourous to include both income and employer status, the size of the excluded group shrinks (see Figure V.2-III) to 6.1% of the self-employed and only 1.0% of all employed people. But, as the final pie charts indicate (Figures V.2-IV and V.2-V), only a very small proportion of employed people, whether they are self-employed, employers, or employees, have annual incomes that exceed $90,000 a year.

Click on image to enlarge

Just as it is no longer sensible to speak of employees or employers as unified categories, so too it must be recognized that labour regulation occurs at different levels and is multi-dimensional, serving a variety of instrumental goals and responding to different normative concerns. Freedland (1995, 24) identifies two levels at which norm making or contracting takes places within work relationships. The executory level is concerned with the basic exchange of services for remuneration, while the relational level is more focused on security of expectations. For workers these are expectations about income and employment security and occupational health and safety, while for employers they centre on how to insure that workers’ activities will help achieve their goals. With a view to creating categories that reflect current socio-legal practice and that have common characteristics in respect of their extension to categories of workers, some scholars have attempted to identify these dimensions (Davidov 2002 a, 67-82; Engblom 2001). Samuel Engblom (2001, 217-20) has suggested that labour law plays at least three roles: the protection of human rights; the promotion of social justice in regard to workers’ economic dependence and subordination to managerial authority; and as an instrument of economic policy to increase the total amount of and change the character of goods and services produced.

Drawing on these ideas, it is useful to identify three dimensions of labour regulation (see Table V.1 below). First, there is a social justice dimension, which includes human rights concerns (for example, anti-discrimination law and pay and employment equity) and occupational health and safety regulation. These laws operate at the relational level of contracting and have been grouped together because they primarily respond to widely held social norms to the partial exclusion (at least normatively and rhetorically if not practically) of economic concerns. While Engblom (2001) includes under the rubric of social justice concerns matters related to economic unfairness and the democratic deficit in work relations, this view is not widely accepted in the North American context. For this reason it is necessary to discuss regulations designed to address economic unfairness and the democratic deficit in relation to other norms as well as in relation to social justice (Davidov 2002 a, b). Thus these concerns have been placed under the rubric of economic exchange and governance, which includes employment standards and certain common law duties such as implied notice upon termination, as well as collective bargaining. However, because this grouping operates at both the executory and relational levels of contracting, its components might require separate treatment. For example, maternity and parental leave and payment provisions contained in minimum standards and employment insurance laws are crucial if women are to enjoy equality in a market economy, and so should be and, in part, are recognized as having an important human rights dimension. In contrast, collective bargaining, at least in the North American context, operates much more at the relational level of contracting and is influenced by diverse norms, including instrumental goals of economic policy (Glasbeek 1987). The third dimension of labour regulation encompasses social wage and social revenue, including employment insurance, public pensions, workers’ compensation, and income tax. Although historically under-developed in the North American context, social wages and revenue are vital both to workers’ well being and to the operation of labour markets, since they provide a baseline on which labour-market exchange occurs. Operating at the relational level of contracting, this dimension of labour regulation is very much influenced by often competing claims of social justice and economic policy considerations.

Table V.1

Dimensions of Labour Regulation

Having identified types of contracts for the performance of work and dimensions of labour regulation, there is now the more difficult challenge of making recommendations about the appropriate extension of each dimension to labour regulation to the different types of contracts. However, before proceeding to these recommendations, it will be useful to first map the current situation in Canada in relation to the conventional typology of work relations despite the difficulties with these categories noted in Part Four. This table (V.2) can also be compared with Figure IV.1, which depicts the personal scope of labour regulation in Canada.

Table V.2

Personal Scope of Labour Regulation - Present

The starting point for our recommendations is that all dimensions of labour regulation should be extended to all workers, defined as persons economically dependent on the sale of their capacity to work, unless there is a compelling reason for not doing so. If this definition were applied to the data on self-employment presented in Part Two, most independent contractors would be included within the personal scope of labour protection. The only group that should be excluded are people who are not dependent upon the sale of their capacity to work. However, such an extension of the personal scope of labour law will require institutional redesign that takes into account the different social and economic relations of different groups of workers. As a result, some of the existing distinctions between different employment statuses will have to be retained for the purpose of designing different regimes of labour protection even though the distinctions are irrelevant for the purposes of determining the scope of labour law protection.

Our starting position is justified on normative, economic, and institutional grounds. The normative rationales are fairness and social justice. Fairness requires that similarly situated individuals be treated equally. In this context, people who are dependent on the sale of their labour prima facie should be treated similarly regardless of the legal form that the transaction takes. The social justice argument is that all workers prima facie should be protected against harms and risks that are broadly seen by society as unacceptable. Moreover, as the expert report for the European Commission noted, the current restrictive approach to labour protection may have discriminatory impacts upon women, exposing them to unacceptable risks:

It may be particularly detrimental to women to restrict the scope of application of labour law and its main guarantees to the field of subordinate employment and the traditional contractual form of such employment, namely the employment contract, without taking account of work performed for others that is channelled through other kinds of legal or contractual relations: known as independent, autonomous, or self-employment or similar. The continued identification of labour law with the regulation of the prototype of labour relations associated with the industrial model that gave rise to such relations – which, moreover, was never fully representative even of all dependent or subordinate work – limits the protection afforded to a smaller and smaller core of workers and leads to even greater segmentation of the labour market (Supiot 2001, 180-1).

The economic rationale for an expansive approach to the personal scope of labour regulation is not only a matter of efficiency, but is tied to the goal of economic activity – improving living standards. According to economist Joseph Stiglitz (2002, 20), “if improving living standards is the objective of economics, then improving the welfare of workers becomes an end in itself; and only if one believes that the market leads to efficient outcomes can one feel confident in not paying explicit attention to workers’ welfare, trusting that the market will make all the correct trade-offs.” Stiglitz has identified a number of information imperfections that distort the labour market. Maintaining the distinction between employees and independent contractors may also produce distortions. Differences in treatment of workers on the basis of differences in the contractual form of their work relationship will induce the stronger party – typically the purchaser of labour – to create less regulated commercial relationships over more regulated employment relationships. This process will create a downward pressure on wages and working conditions and may not lead to compensating improvements in the living standards. Thus, expanding the personal scope of labour protection is an important element of economic policy

The institutional rationale for extending labour regulation to all contracts for the performance of work is that drawing and policing boundaries between different categories of workers has proven to be extremely difficult. For example, as seen in Part Four, even where the personal scope of labour regulation was extended to dependent contractors, as in the case of collective bargaining, the boundary between dependent and independent contractors has proven to be a source of considerable difficulty for labour boards vested with the responsibility for patrolling it. The elimination of distinctions between different categories of workers would lessen the administrative burden, and reduce the uncertainty that inevitably arises in regimes where these distinctions are salient. Boundary problems, however, would not disappear even if the distinction among categories of workers were abolished, for it would still be necessary to distinguish between workers and product sellers. Tests based on the amount of business capital, income, and/or the number of workers under contract would distinguish between workers who depend on selling their services in order to survive and entrepreneurs who invest capital and hire workers. Such tests also have the advantage of being administratively feasible, although they are not immune to manipulation through shams and other attempts to avoid regulation.

As indicated, however, this broad approach to the personal scope of labour regulation merely represents a starting point. There may be compelling normative and economic arguments for altering the personal scope of various dimensions of labour law, either to narrow it to only some categories of workers, or to extend it to persons who are not workers (for example, dependent product sellers) according to our preliminary definition. There is also the practical problem of institutionalizing the principle that all people who perform services for remuneration should be covered by labour protection. In Canada the regimes of labour regulation are, by and large, designed for traditional employees. Until new forms of labour protection can be devised that do not hinge on a traditional employment relationship, it may be necessary, in some circumstances, to retain for the time being the distinction between independent contractors and employees. However, the starting premise is that the distinction between independent contractors and employees is neither a principled nor practical basis for determining the scope of labour protection.

The following sections consider in more detail arguments specific to the different dimensions of labour regulation and identify the kinds of modifications that would have to be made if the domain of labour regulation were extended beyond its present boundaries. It also identifies some of the problems with institutionalizing recommendations that address the issue of personal scope without also exploring issues relating to the delivery of benefits and the design of instruments. However, no attempt is made to provide a highly detailed set of recommendations in respect of those modifications.

As demonstrated in Part Four, because of the normative foundation on which the social justice dimension of labour regulation rests, it has been extended most broadly, particularly in respect of anti-discrimination law. There is no normative or economic justification for permitting discrimination against any worker who provides service personally, regardless of the contractual arrangement under which the service is provided or the form of remuneration stipulated (Davidov 2002 a, 31; Davies and Freedland 2000a; Maltby and Yamada 1997; Perritt 1988). Moreover, this is also a situation where it is appropriate to extend labour regulation into the realm of transactions between product sellers for it is no more acceptable for an auto manufacturing company, for example, to refuse to purchase auto parts from a black-owned business than it is for it to refuse to contract with a black independent or dependent contractor or to refuse to hire a black employee (Davies and Freedland 2000a).

With respect to pay and employment equity, the normative case for extending their application to all workers is also compelling. There is no reason why independent contract work performed predominantly by women should not be as well paid as independent contract work of equal value performed predominantly by men. Nor for that matter is there a compelling normative argument for not requiring firms to review their contracting systems to remove barriers adversely affecting groups of contractors and product sellers that have historically been disadvantaged and to set targets so that these groups obtain a proportionate share of the value of work and product contracts. However, current pay and employment equity regulations and statutes are extremely complex and they tend to be designed to accommodate the existing institutions and practices of collective bargaining. Detailed studies are required to determine what institutional and administrative arrangements would need to be made in order for the pay and employment equity regimes to operate effectively outside of the traditional employment relation.

The case of occupational health and safety regulation is also complicated because it combines direct regulation with participatory rights for workers. As shown in Part Four, many jurisdictions already impose duties on employers to protect all workers with whom they contract, regardless of whether they are employees, dependent or independent contractors. All Canadian jurisdictions should adopt this approach since there is no reason why independent contractors should be exposed to hazards to which employees cannot be exposed. Moreover, independent contractors should be under a duty to obey occupational health and safety laws as they apply to their work (Davies and Freedland 2000a; Perritt 1988). There is no justification for allowing self-exploitation of a kind that is unacceptable in the employment context. This position already been taken in several jurisdictions in Canada.....

Similarly, some participatory rights, such as the right to know about hazardous conditions present at the workplace should, and already do, apply to all workers. The right to refuse unsafe work should also be extended to all workers, including independent contractors; however, some modifications may be necessary to suit the particular circumstances of independent contractors. Where the hazardous condition that causes the work refusal is one that is created by the employer, there is no normative or economic justification to require independent contractors to face the loss of their contracts because they reasonably believe that its performance would endanger themselves or other workers. The procedures created to resolve work refusals by employees would also be suitable in these circumstances for independent contractors. However, where the hazard is created by an independent contractor or is inherent in the work that was contracted (and was known to be so), the work clearly should stop but cancellation of the contract might be permitted or, indeed, required. Finally, rights to representation on joint health and safety committees should be extended to independent contractors in circumstances where the duration of the contract with a particular employer exceeds some minimum (for example, three months). Again, there is no normative reason why independent contractors should not be able to participate in the internal responsibility system of the employer; the only issue is one of administrative and institutional practicality.

Finally, it is worth considering whether occupational health and safety regulation should be extended to both dependent and independent product sellers. The law applies to them as employers in relation to their own workers; the question here is whether it should extend to their contractual relations with other product sellers. In fact, there are some ways in which the law already regulates these relationships. For example, right to know laws require importers and manufacturers to label hazardous products and to provide material safety data sheets to purchasers in order to insure that information is transmitted with the product and reaches its end users. As well, when it comes to contracted-in services, the law draws no distinction between independent contractors and product sellers; when one firm contracts for services from another firm there is overlapping responsibility between the two firms for the health and safety of the workers of the contracted-in firm performing the work. This overlapping responsibility creates an incentive for contracting-in firms to exercise due diligence in regard to the occupational health and safety practices of the firms and individuals with whom they contract. In this same spirit, it has been suggested (Gunningham 1998, 228) that bidders on government contracts over a certain value should be required to certify that they have a safety management system. Consideration should be given to other ways of intervening in supply chains to improve health and safety performance. Specifically, the possibility of placing firms under a general duty to contract with firms that operate in compliance with applicable occupational health and safety laws, subject to a defence of due diligence should be investigated. Further study should be undertaken to develop more specific proposals to implement this approach.

Employment and labour regulation that is directed at economic terms and the governance of work relations balances the goal of social welfare for working people and promoting efficiency in the labour market (Davies and Freedland 1999, 233). However, it is important not to assume that these goals are either uniformly consistent or inconsistent; both economic theory and empirical data suggest that the relationship between them is much more complex (Addison and Siebert 1997; Deakin and Wilkinson 2000; Kitson, Martin and Wilkinson 2001). But as Davies and Freedland (1999, 247) acknowledge

it is unlikely that real progress will be made in refining the debate about the personal scope of employment law until ways can be devised of interrelating, on the one hand, the functional needs for the sort of regulation offered by particular aspects of employment law, with, on the other hand, the complex incentive structure and allocation of risks which are embodied in particular work arrangements.

Currently, minimum standards legislation generally applies only to traditional employees, although through adjudication its personal scope has sometimes been extended to dependent contractors. At the very least, minimum standards should be expressly extended by legislation to dependent contractors. However, the normative case for extending minimum standards to all workers who are economically dependent on the sale of their labour (that is, all but product sellers) is a strong one. If we do not allow competition between traditional employees and, perhaps, dependent contractors to produce economic terms and conditions that fall below socially acceptable levels, why should we allow such outcomes among independent contractors, especially in a context where the boundary between the two groups is particularly elusive and our data show that a majority of independent contractors do not possess the distinctive features of entrepreneurship – ownership, autonomy, or control over production.

Numerous difficulties, however, may arise in the application of the principle that all people who personally perform work for remuneration should be covered by minimum standards legislation.[120] For example, the difference in control exercised by the employer may make the comparison unworkable. In the case of traditional employees and dependent contractors, the control exercised by the employer over the performance of work makes it possible to speak meaningfully about a minimum wage for a specified period of time. However, in the case of independent contractors who have been engaged to complete a task for a price, the lack of control exercised by the employer makes it impossible to calculate a minimum wage since the effort level is largely within the control of the independent contractor. A second but related concern is that while an employee is paid only wages, the price paid to an independent contractor includes, in addition to remuneration for the contractor’s labour, the cost of the contractor’s own employees, the supplies and materials used by the contractor, and the use of the contractor’s capital. A third concern is that because independent contractors are, almost by definition, unlikely to be dependent on a particular employer but rather to be hired by numerous employers, problems are likely to arise in respect of such matters as hours of work and overtime provisions. These concerns suggest that it is important to consider the type of standard when determining the specific scope of protection. Income and hours of work thresholds are likely more appropriate than employment status for establishing the scope of specific standards.

In addition to these difficulties, it is also necessary to accommodate the great variation in the way different economic and industrial sectors are organized. For this reason, much of the regulation directed at economic terms and governance structures should be specific to a sector. For example, the construction, garment, and trucking industries have distinctive contractual relations and employment structures and a history of distinctive regulation (ILGWU and Intercede 1993). It may also be the case that labour principles will be brought to the design of special protective laws for independent contractors (Davies and Freedland 2000a, 43-4). In sum, because an extension of minimum standards to independent contractors raises serious concerns and design problems, further study should be undertaken to investigate these problems, determine their magnitude, and explore possible solutions.

Finally, there are good reasons for the partial extension of minimum standards legislation to product sellers. First, all product sellers should be jointly and severally liable for the failure of the parties with whom they contract to adhere to minimum standards, subject to a due diligence defence. This approach would parallel the previous recommendation with respect to occupational health and safety regulation. There is ample evidence that chains of sub-contracting often produce situations in which workers are employed at the bottom end of the chain by transient or insolvent entities that do not adhere to minimum standards and are effectively immune from enforcement actions (Becker 1996; Collins 1990; ILGWU and Intercede 1993). In the absence of some element of common management or control, employees cannot hold contracting entities higher up on the chain jointly and severally liable as related employers under employment standards legislation.[121] This recommendation would help protect employees against the harmful effects of these kinds of arrangements. Second, there are some settings, such as the fishing industry, where product purchasers have organized production in a manner that has off-loaded risks to small producers, in a context where those producers lack the economic leverage to protect themselves against exploitive conditions. In these circumstances, it is appropriate to specifically designate these dependent product sellers as workers even though they would fall outside the category of “workers” because of the amount of capital they own or the number of workers they hire. Alternatively, specialized minimum standards regimes could be created for particular groups of dependent producers geared to the particular conditions in the industry.

In general, collective bargaining schemes have already been extended to dependent contractors. The question that needs to be addressed is whether they should be further extended to all workers, including independent contractors and, in special cases, dependent product sellers. The ILO’s Convention concerning Freedom of Association and Protection of the Right to Organize (Convention 87 of 1948) guarantees the right of “workers and employers, without distinction whatsoever” to establish and join organizations of their own choosing without state authorization (Benjamin 2002, 80). According to the Freedom of Association Committee of the ILO’s governing body, self-employed workers in general should enjoy the right to organize and the existence of an employment contract should not determine whether a person is covered by the right (ILO 1996, 51). Therefore, the starting position should be that this dimension of labour regulation covers all people who perform work personally unless there are strong normative, economic, or institutional reasons for excluding them. The most obvious concern is that collective bargaining collides with competition law in that it allows combinations among sellers of labour power that are not normally permitted to product sellers. The reason for this different treatment is that labour is not an ordinary commodity since it cannot be separated from the workers who sell it and that individual workers commonly lack adequate bargaining power to negotiate socially adequate economic terms or obtain a voice in workplace governance. However, as Pietro Ichino (2001, 191) notes “the legal distinction between the (unlawful) restriction of competition in goods and services market and the (lawful) restriction of competition in labour market is only theoretically clear-cut.”

Independent contractors do not fit neatly into the justifications for collective bargaining (Davidov 2002 a, chapter 2). While they are dependent on the sale of their labour power, although they may not be economically dependent on any particular employer, it is an open question as to whether or not they lack adequate bargaining power as individuals to obtain socially adequate outcomes. Moreover, some independent contractors may be competing with employees, while others may not. The statistical portrait of the self-employed provided in Part Two indicates that independent contractors are a diverse group - some do very well (such as independent professionals like lawyers and accountants), while others struggle to make ends meet. This concern about over-inclusiveness can be met by making provision for the exclusion of groups for whom it is determined that collective bargaining is both unnecessary and unduly interferes with other public policy objectives, as it currently done in collective bargaining statutes (for instance, the exclusion of professionals in Ontario). Another technique would be to draw upon competition law concepts such as product differentiation and barriers to entry in order to distinguish between independent contractors who should be covered by labour law and those who are more appropriately placed in the commercial realm (Perritt 1988, 1040-41). A second concern is that the wholesale exclusion of product sellers may exclude groups, like fishers, who should be given access to collective bargaining because of their economically dependent condition. To meet this concern about under-inclusiveness, provision should be made to allow dependent product sellers to be designated as workers for the purposes of collective bargaining.

Merely giving independent contractors and designated dependent product sellers access to collective bargaining, however, might not accomplish very much, given the high likelihood that they work for multiple employers under fixed- and short-term contracts. The basic collective bargaining scheme is not suited for these conditions (Fudge and Tucker 2001). Therefore, alternative models, such as those developed for some sectors of the construction industry (ILGWU and Intercede 1993) and for artists and performers (Fudge and Vosko 2001; Langille and Davidov 1999; MacPherson 1999) will need to be examined. Further study should be undertaken to investigate the characteristics of independent contractors and dependent product sellers with a view to developing appropriate collective bargaining mechanisms.

According to the ILO (2000a, 18) “social protection is not only morally indispensable but also economically viable. An efficient economy and an effective system of social protection are both essential for the attainment of income security and a stable society.” It also declared “extending personal coverage is probably the greatest challenge facing social protection systems” (ILO 2000a,14).

Unlike other dimensions of labour law that aim specifically to regulate the conditions under which work is performed, social wage and social revenue are fundamentally mechanisms for providing for the economic welfare of citizens. It is an historic choice to link access to and funding for these benefits to the employment relation (Fudge and Vosko 2001b; Langille 2002). A system of universal entitlements would protect the economic welfare of all citizens, regardless of their attachment to the labour market.

The Supiot report to the European Commission addressed the issue of extending social protection at the same time as it acknowledged the need to attend to economic issues (Supiot 2001, chapter 2). It argued for a reconsideration of the notion of security along three lines (Deakin 2002, 189; Supiot 2001). The first involves focusing on the concept of labour market status (referred to as ‘”statut professionel” in the report), which emphasizes labour market participation over an individual’s life cycle rather than employment status per se. The second concentrates on an extended concept of work in place of the narrow notion of “employment” as the basis for access to social rights and protections. The third introduces the idea of “social drawing rights” which individuals can use to achieve security with flexibility in conditions of uncertainty. These proposals suggest a basis for reconsidering the scope of social protection that would loosen or de-link altogether entitlement from labour force participation and employment status.

However, within the confines of the social wage regime that is built around labour force participation, our starting point is that all workers should be protected against the risk of loss of the ability to sell their capacity to work, whether due to lack of available buyers, injury, sickness, or old age regardless of whether they have a contract of employment. In short, employment insurance, public pensions, and workers’ compensation should be available to all workers, regardless of the type of income they receive in return for the provision of service.

Several jurisdictions already allow self-employed workers voluntarily to self-insure, as in the case of workers’ compensation. At the very least, this opportunity should be provided to all self-employed workers for all social protection schemes. Indeed, this has been the formal position of the ILO since 1944 when it adopted the Income Security Recommendation, 1944 (No. 67), which called for the insurance of self-employed persons “against the contingencies of invalidity, old age and death under the same conditions as employed persons as soon as the collection of their contributions can be organized (ILO 2000a, 197-8).” As well it recommended that consideration be given to insuring them against loss of income due to sickness and maternity.

But even this arrangement fails adequately to address the underlying inequity that arises by the requirement that self-employed persons pay the full premium personally, while persons who are classified as employees have their employers pay all or part of their contributions, depending on the scheme. There is no reason why workers should have to demonstrate that their income was earned under a contract of employment or under conditions of economic dependency on or subordination to a particular employer in order gain the benefit of their contributions.

The recommendations listed below address the question of the personal scope of labour protection from the perspective of labour law policy and labour law institutions. The recommendations do not address technical questions relating to drafting specific definitions nor questions relating to statutory interpretation.

1. Extend the personal scope of labour law to all workers, defined as persons dependent on the sale of their labour, unless a compelling reason can be provided for the exclusion of a well-defined sub-group.

2. Relevant distinctions between different groups of workers (such as the nature of their relationship to the entity that purchases their service) should be taken into account in the design of instruments that provide labour protection to all workers, regardless of the type of income they receive in return for the provision of service.

3. It may be appropriate in some cases to extend the scope of labour law to contracts between product sellers for the purpose of protecting dependent product sellers and workers.

1. Social Justice

a) Anti-Discrimination Law

4. Extend anti-discrimination law to all contracts for the performance of work and to the purchase of goods and services by contractors and product sellers.

b) Pay and Employment Equity

5. Investigate the design of a pay equity scheme that applies to all work contracts.

6. Investigate the design of a scheme that would extend the principles of employment equity (for instance, system reviews and targets) to all work and product contracts.

c) Occupational Health and Safety

7. Impose on employers the duty to provide healthy and safe working conditions for all workers with whom they contract.

8. Impose on contractors the duty to obey applicable health and safety laws in the performance of their work.

9. Investigate the design of occupational health and safety laws that extend to all workers the right to know about hazards and to be trained in minimizing such hazards, the right to participate in the design and implementation of safe and healthy work systems, and the right to refuse unsafe work.

10. Investigate the design of an occupational health and safety law that imposes a duty on product sellers to contract only with other product sellers who comply with occupational health and safety laws, subject to a due diligence defence.

2. Economic Terms and Governance

a) Minimum Standards

11. Extend the scope of minimum standards to dependent contractors as defined under current collective bargaining law.

12. Investigate the design of a minimum standards law that applies to all workers. Consider sectoral-specific regulation to capture the distinctive features of employment relations in certain industries.

13. Provide for the administrative designation of some groups of dependent product sellers as workers covered by all or part of minimum standards laws, including the power administratively to designate their employer.

14. Investigate the design of a law that would make product sellers jointly and severally liable to workers of other product sellers in the production chain for money owed under minimum standards law.

b) Collective Bargaining

15. Investigate the design of a collective bargaining scheme suitable for all workers, including so-called independent contractors, who have multiple employers. Consider sectoral-specific regulation for certain distinctive industries.

16. Provide for the administrative exclusion of groups of workers where a strong case can be made that, based on their individual bargaining power, collective bargaining is both unnecessary and would unduly interfere with other public policy objectives (for example, competition concerns).

17. Provide for the administrative designation of some groups of dependent product sellers as workers covered by collective bargaining law, including the power to designate their employer.

c) Common Law

18. Provide an implied right to notice of termination for all workers under contracts of indefinite duration.

3. Social Wage and Social Revenue

19. Investigate the design of a system of universal entitlements that would protect the economic welfare of all citizens, regardless of their attachment to the labour market, funded from general revenue.

20. While social protection programs remain constructed around labour force participation, investigate the design of a system that covers all workers and that does not require workers not employed under a contract of employment to make greater contributions than employed workers.