A Comparison of Labour Standards in the United States and Canada.

This article introduces a methodology for measuring differences in the labour standards between the United States and Canada, taking into account variations by state and province. This methodology is then used to analyze differences in the two countries on ten labour standards. The results indicate

that six standards are higher in Canada than in the United States: paid time off unemployment/employment insurance, workers' compensation, collective bargaining, unjust discharge and advance notice of plant closings/large scale layoffs. Standards covering minimum wages, overtime and occupational safety and health are higher in the United States than in Canada. There is no difference in the two countries in standards covering employment discrimination/employment equity. The results suggest that overall, although there are exceptions, labour standards are higher in Canada than the United States.

The 1993 North American Free Trade Agreement (NAFTA), in combination with the 1988 U.S.-Canada Free Trade Agreement (FTA), has raised with new urgency issues associated with free trade between the United States and Canada. The open trade encouraged by NAFTA and FTA is expected to cause increased competition between firms in both countries, raising concerns about how workplace standards will be affected. [1] As competition and international trade increase, firms are likely to view the maintenance of quality employment relations as affecting their competitive advantage in the product market. At the same time, employees who may view themselves as subject to the adverse effects of competition may see the maintenance of labour standards as an important component of their welfare. [2]

Turning to the United States and Canada, the traditional view is that U.S. labour standards are lower than Canadian labour standards (Burton 1989; Martin 1991). Assuming that the traditional view is accurate, two basic, but interrelated concerns exist (Compa 1993). If lower labour standards in the U.S. result in lower costs of production for U.S. firms vis-[grave{a}]-vis Canadian firms, the prices of goods for U.S. firms will be lower than the prices of comparable goods sold by Canadian firms, other things being equal. One possible result will be that Canadian firms will lose market share and reduce employment; in the extreme case, these firms would simply go out of business, causing still greater unemployment. A second possibility is that Canadian producers will shift production to the U.S., either through investment or subcontracting. In such a situation, while Canadian firms would continue to prosper, Canadian workers would experience unemployment (Addison, Fox and Ruhm 1995). Under any of these scenarios , Canadian workers would experience greater unemployment.

While it is generally presumed that U.S. labour standards are lower than Canadian labour standards (Block 1994, 1996; Card and Freeman 1994; Hyatt and Kralj 1992; Roberts and Madden 1992; Gunderson 1998), the validity of this presumption has never been tested and differences in labour standards between the two countries have yet to be comprehensively analyzed. The purpose of this research is to begin to fill this empirical gap in our knowledge of Canadian and U.S. labour standards. The ultimate objective of this project is to develop an understanding of the differences between United States and Canadian labour standards and to develop a method for operationalizing differences in labour standards. This will permit researchers to include measures of labour standards in future analyses of trade effects, thus permitting policymakers in the U.S. and Canada to draw on this research to inform their decisions regarding trade and labour standards.

To that end, the second section of this paper will discuss the overall context for consideration of labour standards in the U.S. and Canada. The third section discusses the methodology used to make the comparison. The fourth and fifth sections will discuss our results and the sixth section will provide a

summary of our work and some conclusions and implications.

OVERVIEW OF THE DEBATE

The United States and Canada are neighbours and one another's largest trading partner. Both are developed western countries that trace their origins to British rule in the 18th century. Neither nation had its economic infrastructure damaged or destroyed during World War II; thus the plant and equipment in the two countries have followed comparable development cycles. Many firms operate on both sides of the border (Lipset 1989; Rugman 1991).

Despite these similarities, there are important divergences in views about the relationship between the individual and the government and the role of the state. The United States is generally considered anti-statist and individualistic (Lipset 1989; Blank 1994; Block 1992). Its entire constitutional and governmental structure is built around limiting the power and role of government. Canada, on the other hand, is more statist than the United States and has traditionally been more willing to accept some governmental control over the lives of its citizens in order to obtain security and order (White 1988; Lipset 1989).

This difference in values between the U.S. and Canada toward the role of the state was manifested most clearly in the debate around the 1988 Canada-U.S. Free Trade Agreement. The FTA generated far more public debate in Canada than the in U.S. (Mahant 1993). Canadian opponents of the FTA made their arguments in terms of sovereignty, expressing the view that the Canadian commitment to a high level of social welfare for all its citizens would be compromised if free trade resulted in pressure to harmonize downward its policies with U.S. policies (Lyon 1987; Doern and Tomlin 1991; Martin 1991; Mahant 1993; Smith 1988; Gunderson 1998).

A key theme in this debate was the perceived Canadian commitment to social welfare in the form of high labour standards. [3] The concern on the part of Canadian FTA opponents seemed to be that in order to assure competitiveness in the product market, Canadian firms would use the pressure of free trade as a political lever to argue for reduced labour standards obligations, would reduce compensation to offset the costs of the higher Canadian labour standards and/or move production to a location that was believed to have less burdensome and, therefore, less costly, labour standards. In all cases, the welfare of workers/citizens would be reduced, through a reduction in standards, through a reduction in compensation and/or through a reduction in employment.

This view was based on the assumption that Canada's labour standards are superior to those of the United States, that the difference is substantive and that, in a free trade environment, these higher labour standards would put Canadian firms at a disadvantage relative to their U.S. counterparts, other things (such as exchange rates) being equal. Empirically, however, the extent to which Canadian labour standards are actually higher than U.S. standards and therefore impose higher costs on Canadian producers vis-[grave{a}]-vis comparable U.S. producers has never been demonstrated. [4] This article tests the proposition that Canadian labour standards are higher than United States labour standards.

METHODOLOGY FOR COMPARING LABOUR STANDARDS

As a first step in the process, it is necessary to develop a definition of "labour standards" that can be commonly applied across the two countries. The key characteristic of a labour standard is that it is applicable to all, or almost all, employers and employees. Thus, our research views government mandates as establishing a minimum, legally enforceable floor for labour standards. A labour standard is defined as any governmentally established procedure, term or condition of employment, or employer requirement that has as its purpose the protection of employees from treatment at the workplace that society considers unfair or unjust. The common element across all standards is that they are mandatory -- they are governmentally imposed and enforced. Employer failure to comply with the standards brings legal sanctions upon the employer.

The Labour Standards Analyzed. Based on this definition, labour standards fall into one of two categories: (1) standards that require employer monetary payments, either to workers or to a government agency; and (2) standards that place constraints on employer actions vis-[grave{a}]-vis workers. The standards analyzed that require employer payments were minimum wage, overtime, paid time off, unemployment/employment insurance and workers' compensation. The standards that placed constraints on employer actions vis-[grave{a}]-vis employees were collective bargaining, non-discrimination (equal employment opportunity in the United States, employment equity in Canada), unjust discharge, occupational safety and health and advance notice of plant closings/large scale layoffs. [5] All of these meet our working definition; all of these involve some government intervention or regulation of the workplace. [6]

Methodological Issues in Canada-U.S. Comparisons

Under ideal circumstances, one would simply determine the labour standards to be analyze and then examine these for the United States and Canada. There are, however, two difficulties with this approach. First, different levels of government promulgate labour standards in the two countries. For example, while collective bargaining is federally regulated in the United States, it is regulated by both the provinces and the federal government in Canada. This is likely to be the case across most of the standards. Thus, any analysis of labour standards in the United States and Canada must take account of the varying levels of government and jurisdictions that promulgate labour standards.

Second, a critical difference between the U.S. and Canadian governmental structures is that in the U.S., if there is a federal statute governing some aspect of the workplace, the federal rule prevails unless the state statute raises the standards. In some cases, states will have statutes that appear to allow lower standards. Such state statutory standards apply to classes of workers excluded from federal legislation, typically those that do not affect interstate commerce. In contrast, with a few exceptions, the federal standard in Canada usually only applies to federal government employees and those industries that can reasonably be thought of as involved in inter-provincial commerce (Government of Canada 1988). Except for employees in those industries, the provincial standards prevail.

Methodology for Comparing Labour Standards. Our methodology for comparing labour standards involves three components: (1) an analysis of the substance of each of the standards, including the actual statutory provisions and the nature of enforcement; (2) development of an index of the strength of the labour standard in a jurisdiction through the use of a weighting scheme for the various statutory provisions composing each of the standards; and (3) deflating the weighted standards by an estimate of the percentage of the labour force covered by each standard.

The first two components provide a measure of the rankings of labour standards as they affect the typical employer or worker in the jurisdiction. They comprise what we will call the basic index. A basic index is created for each labour standard for each subnational jurisdiction (e.g., states in the U.S. and provinces and territories in Canada). To make cross-country comparisons, we generate both an unweighted average for each labour standard for each country as well as one weighted by each subnational jurisdiction's share of its country's employment. These are referred to as the unweighted and weighted indices.

Deflating by the percentage of the labour force covered by each standard creates, for standards for which coverage is relevant, a measure of the overall employee welfare associated with the labour standards. As such, it provides more of an aggregate, societal-level measure than the basic index. This is called the deflated index. Because both the basic index and the deflated index present useful and complementary information, we will present both.

The Substance of the Standards. The main component of the index was based on the statutory substance of the labour standards. This was done through coding and arraying provision categories that were comparable across all jurisdictions. [7] The provisions incorporated into the analysis are discussed in the presentation of the results for each standard. [8] In addition to the substance of the statutory provisions, we incorporated an enforcement component, based on rights of judicial appeal outside of the administrative agency that has primary responsibility for enforcing the statute.9 The broader the rights of appeal from the decision of the administrative agency, the weaker the enforcement mechanism, as the agency is more likely to have an interest in enforcing the statute than the courts. The literature on appeals of administrative decisions indicates that, due substantially to the presence of privative clauses in Canadian labour statutes, appeals rights are broader in the United States than in Canada (Crowl ey 1987; Novak and Somerlot 1990; Mullan 1993; Block 1994; Adams l995).[10]

Development of an Index. The second step in the development of the index was the creation of an index for each of the standards. The index for each standard has two components: a subindex for the provision that is greater the higher the level of protection given to employees; and a weight given to each provision within each standard.

Based on a technique developed for coding collective bargaining agreements (Kochan and Block 1977; Block 1978a, 1978b), an ordinal scale was constructed for each provision. Values were assigned to each relevant statutory provision or enforcement mechanism by assigning a score of zero (0) to the absence of a provision, and a score often (10) to the strongest provision. Provisions of intermediate strength were assigned intermediate values in accordance with the number of possible categories in the provision. [11]

For example, for collective bargaining laws, jurisdictions in which union recognition could be obtained without an election were assigned a value of 10 and jurisdictions in which an election was required were assigned a value of 0. Advance notice requirements for large-scale layoffs represent an example of a provision for which an intermediate coding was used. If the provision of the statute in the jurisdiction required advance notice of greater than or equal to 16 weeks, the jurisdiction was coded as a 10. Notice of 12 to 16 weeks was coded as 7.5, 8 to 12 weeks notice was coded as 5.0, 4 to 8 weeks as 2.5 and no provision was coded as zero.

In addition to coding each provision, a weighting scheme was established for provisions within a labour standard, such that the total weights of all the provisions within a standard equalled 1. Greater weights were given to provisions within each standard that were deemed to be most important. For example, in the minimum wage standard, the level of the minimum wage was weighted .92; the availability of a learner wage was weighted at .04, the possibility of a fine or imprisonment was weighted at .02 and the right of appeal was weighted at .02. [12]

Estimating Coverage

The third component in the full index is the extent to which the labour force is covered by each labour standard. The justification for including coverage is that the proportion of the workforce intended to enjoy the protection afforded by any given labour standard is one component of that standard's effect. Coverage is measured as the proportion of the workforce covered by each labour standard. If all workers within a jurisdiction are covered by a particular labour standard, that proportion is equal to one. What we refer to as the "coverage-deflated" (or simply the "deflated") index is constructed by multiplying the basic index by the coverage proportion. The employment-weighted deflated index is constructed by multiplying the employment-weighted index by the coverage proportion. Incorporation of coverage will lower the indices to the extent that coverage is less than comprehensive.

Coverage criteria are specified in the enabling legislation for each standard. Across the ten standards, there are three bases for coverage criteria: characteristics of the workforce, firm size and occupation/industry. Coverage was assumed to be comprehensive for six of the ten standards. Two have universal coverage in both the U.S. [13] and Canada (protection from unjust discharge and paid time-off [14]) and, therefore, coverage was judged to be equal to one for those two standards. Anti-discrimination laws protect certain classes of individuals, for example according to race or gender. Because of the difficulty in applying several of the coverage classifications, specifically, disability, religion, national origin and, in Canada, political beliefs, we determined that useful coverage estimates could not be generated for anti-discrimination/employment equity laws. For the purpose of constructing the deflated indices, coverage was assumed to be equal to 1. It should be recognized that this assumption introduc es an upward bias in the index, although it does not change the rankings across jurisdictions. Coverage for three of the standards, unemployment insurance, health and safety and advance notice, is based primarily on firm size. In most cases, these exclusions are for very small firms and the necessary firm size data were not available. As a result, full coverage was again assumed to be comprehensive.

Coverage estimates were generated for the four remaining standards, minimum wage, workers' compensation, overtime pay and collective bargaining using data from two sources, the Current Population Survey (CPS) for the United States and unpublished data from the Survey of Employment, Payrolls, and Hours for Canada, both for l993. [15] The CPS data used here were from the March 1993 survey and were read from the "Current Population Survey, March 1991-1993" CD-ROM. The CPS sample is composed of households and includes data at the household and individual levels. Individual records were read and aggregated to construct the coverage estimates used in this study. Annual average employment and union membership data from the monthly Canadian employment survey were provided to us at the 3-digit Canadian 1980 SIC level in tabular form for all of Canada and by province from the Labour Division of Statistics Canada. [16]

Based on the foregoing discussion, the index construction can be generally expressed as follows:

let [S.sub.pdj] = the score assigned to provision p in standard d in jurisdiction j, where 0 [leq] [S.sub.pdj] [leq] 10;

let [W.sub.pdj] = the weight assigned to provision p in standard d in jurisdiction j, where 0 [leq] [W.sub.pdj] 1; and let [X.sub.dj] = the basic index score for standard d for jurisdiction j. Then

[X.sub.dj] = [[[sum].sup.n].sub.P = 1] [S.sub.pdj] * [W.sub.pdj] where the index consists of n provisions.

Let [C.sub.dj] = the coverage deflated index score for standard d for jurisdiction j. Then,

[C.sub.dj] = [X.sub.dj] * [c.sub.dj], where [V.sub.dj] denotes the percentage of employees in jurisdiction j covered by standard d.

COMPARING LABOUR STANDARDS IN THE UNITED STATES AND CANADA

The indices for the United States are summarized in Table 1 and those for Canada in Table 2. The last two columns in each table provide a summing of the basic indices and coverage-deflated indices. [17] This sum provides a sense of the overall level of labour standards in the jurisdiction. The standards used are those in effect as of December 1998.

Table 3 provides a summation of each of the indices for the two countries. The first two rows of Table 3 provide unweighted averages of the indices for the two countries and then a sum of the unweighted averages. The last two rows of Table 3 present the indices weighting the jurisdictions by the percentage of the country's employment in each jurisdiction. [18]

Table 4 presents analogous information for the coverage-deflated indices. Although the differences between the unweighted and weighted indices are not great, the differences for the minimum wage and advance notice standard are noticeable and will be discussed below.

The remainder of this section presents the results for each standard: first, standards that require employer payments; then, standards that place constraints on employer actions vis-[grave{a}]-vis employees. Reader might refer to the tables as each standard is discussed.

Standards that Require Employer Payments

Minimum Wage. Minimum wage levels in effect on December 31 1998, denominated in domestic dollars, are generally higher in Canada than in the U.S. Although seven states and the District of Columbia have enacted minimum wage rates higher than the federal standard of $5.15, the federal standard is clearly the norm. Canadian nominal minimum wage rates range from $4.00 in the federal jurisdiction to $7.15 in British Columbia.

Valued in Canadian dollars, at C$1 = US$.70, however, the U.S. minimum wage would be $7.36, which would place it at the top of the Canadian distribution. The Canadian minimum wage, valued in U.S. dollars, ranges from $2.80 to $5.00. This suggests that the U.S. minimum wage is generally higher than the minimum wage in all Canadian jurisdictions.

This difference is reflected in the basic minimum wage indices (Table 3): 7.24 for the U.S. and 4.05 for Canada for the unweighted Minimum Wage Index and 7.34 and 5.24 for the weighted index. The major reason that the weighted minimum wage index for Canada is greater than the unweighted index is because of the relatively high minimum wage in the three largest provinces, Ontario, Quebec and British Columbia.

Overtime/Hours of Work. Basic overtime standards are higher in the United States than in Canada. While the norm in the United States is a maximum of 40 hours per week, there is far more variation in this norm in Canada. Of the thirteen Canadian jurisdictions, only seven have a forty-hour per week maximum. Of the three largest provinces, British Columbia, Ontario and Quebec, only British Columbia has legislated a 40-hour maximum.

Both countries permit sectoral exemptions, generally for those industries that either do not operate on a regular daily work schedule or for employees who must perform their jobs without supervision and monitoring. Quebec is the only jurisdiction in either country that permits an exception either by collective agreement or with the permission of the government.

The basic index average reflects this difference. For the unweighted indices, the average is 10 for the United States and 7.49 for Canada; for the weighted indices, the average is 10 for the United States and 7.06 for Canada.

Paid-Time Off Although the U.S. generally has more holidays than Canada, there is no requirement in any law or statute in the United States that employees be paid for those holidays. On the other hand, while the Canadian provinces have somewhat fewer holidays than jurisdictions in the United States, mandatory payment to employees for those holidays is almost universal throughout Canada. Every Canadian jurisdiction requires either that employees be paid time-and-one-half or receive the day off with pay.

Unlike the United States, Canadian jurisdictions also have legislatively mandated paid vacations. Mandated vacations range from two weeks to three weeks, with employees receiving 4% or 6% of annual pay. Eligibility commences after a person is employed from ten months to 12 months.

Based on this, it is clear that Canadian standards for paid time off are far higher than comparable U.S. standards. The paid-time-off index reflects this difference. Because we have given the greatest weight to pay entitlement (e.g., twice the weight given to number of holidays in the jurisdiction), the Canadian jurisdictions are associated with basic indices that are substantially higher than the United States indices; 6.37 in Canada, 1.29 in the United States for the unweighted index average; 6.59 and 1.29 for the weighted index average.

Unemployment/Employment Insurance. Our results and indices suggest that the Canadian El system is slightly better, from the workers' point of view, than the United States UI systems. The unweighted Canadian index is 7.51, while the average U.S. index is 6.28. The weighted average indices are similar, 7.51 for Canada and 6.09 for the United States.

This small difference in favour of Canada is due to two factors. First, the average weekly benefit as a percentage of previous earnings in Canada is higher than in the United States. The standard in Canada is 55%, while in the United States the average is 37%. Second, Canadian workers may draw their benefits for a longer period of time than United States workers, 45 weeks as compared to a maximum of approximately 39 weeks (with extended benefits) in the United States. The typical length of time in the United States is 26 weeks.

While the benefit structure to employees is higher in Canada than in the U.S., Canadian workers must directly pay for these higher benefits. Unlike in the U.S., where the total cost of unemployment insurance is borne by employers, the cost of unemployment insurance is shared in Canada. Canadian employees pay $2.95 per $100 of insurable earnings up to $39,000 per year. U.S. workers pay nothing. This somewhat offsets the advantage that unemployed Canadian workers have vis-[grave{a}]-vis their counterparts in the United States. [19]

Workers' Compensation. The criteria used to represent the dimensions of workers' compensation are based on the nineteen essential recommendations from the National Commission on State Workmen's Compensation, published in 1972. Among the recommendations are eight related to coverage, nine related to the adequacy of income benefits and two related to scope and adequacy of medical benefits. Data on compliance across states is compiled by the Employment Standards Administration within the U.S. Department of Labor. [20]

The average unweighted basic indices for the two countries are 6.72 and 7.6 for the United States and Canada, respectively, indicating that the Canadian provinces are somewhat more likely to be in compliance with the nineteen essential recommendations that are the U.S. states. The distribution of scores shows that the highest scoring state, Iowa, actually has a higher score than the highest scoring Canadian jurisdiction, the Northwest Territories. Only four of the Canadian jurisdictions, however, have scores below the U.S. national average, indicating more prevalent compliance with the essential recommendations in Canada than in the United States. Looking at the subcomponents of the essential recommendations, the Canadian laws are more complete with respect to coverage and slightly less complete with respect to benefits levels. Although actual benefit levels tend to be slightly higher in Canada than in the U.S., the Canadian provinces in general provide slightly less extensive benefits to widows and survivors in the event of death.

Standards that Place Constraints on Employer Actions vis-[grave{a}]-vis Employees

Collective Bargaining. Given the enormous amount of legal doctrine on collective bargaining, it was necessary to collapse this area into fundamental concepts. Six doctrines and laws seem to be most important in determining whether laws favour employers or unions: the necessity of an election for union certification, the scope of bargaining, governmental involvement in negotiations, the ability of the employer to permanently replace economic strikers, the availability of first agreement arbitration to protect fledgling unions and coverage (Weiler 1983; Block 1994; Chaison and Rose 1994).

As is generally acknowledged, Canadian labour laws are far more favourable to unions and collective activity than U.S. labour laws. Eight of the eleven Canadian jurisdictions do not require an election for union certification, thus minimizing the role of the employer in the unionization process. None of the Canadian jurisdictions place limits on the scope of bargaining, thus permitting union involvement in a wide range of employer decisions. All Canadian jurisdictions incorporate conciliation during negotiations, either at the initiation of the provincial government or at the request of one party, thus making it more difficult than otherwise for the employer to use the bargaining process to eliminate the union. Two jurisdictions have a broad-based ban on striker replacements and seven jurisdictions permit first agreement arbitration, thus institutionalizing the union.

All these differences result in a substantially higher index for the Canadian jurisdictions than for the United States. The unweighted Canadian average is 7.92, as compared to a score of 1.5 for the United States. The weighted Canadian average is 7.10, while the U.S. average, because it is based on national legislation, does not change. The change in the Canadian index is due to the greater weight associated with Ontario, which has relatively weak collective bargaining legislation, relative to the other provinces.

Equal Employment Opportunity/Employment Equity. The EEO/EI index, which is heavily weighted toward racial and gender discrimination, indicates that this labour standard is roughly equal in the two countries. The relevant unweighted basic index averages are 8.64 in the United States and 8.68 in Canada. The weighted index averages are 8.70 in the United States and 8.53 in Canada. The slight drop for Canada is due to the fact that three of the larger provinces, Ontario, Quebec and Alberta, have index scores slightly lower than the unweighted average.

Canadian law covers a slightly wider range of activities and classes than U.S. law. Laws in both countries cover race/aboriginal peoples/visible minorities, gender, national origin, religion and age. A higher percentage of Canadian provinces than U.S. states cover sexual orientation. In addition, most Canadian jurisdictions protect persons from discrimination based on political beliefs and membership in organizations, protection that is not extended to employees/applicants in the U.S. (with the exception of union membership). U.S. law, on the other hand, goes farther than most Canadian jurisdictions in requiring reasonable accommodation for disabled employees (Schneid 1992). Finally, Canadian laws generally provide for comparable worth between traditionally female and traditionally male occupations. U.S. laws have no such provisions.

A key difference, albeit uncoded, between equal employment opportunity law in the United States and employment equity in Canada, is that the latter is grounded in concepts of human rights comparable to constitutional rights in the U.S., while the former is viewed as a regulation of interstate commerce. In the U.S., there has been a reluctance to permit employees to make constitutional claims against private employers (Kelly 1991; Wolkinson and Block 1996).

Unjust Dismissal. Neither country has broad-based protection against unjust dismissal. In the U.S., only Montana provides statutory protection against unjust dismissal. Roughly forty states have developed judicial doctrine based on contract law that prohibits an employer from discharging an employee except for just cause where the employer has promised not do to so. Such contracts may be written, implied, or based on a handbook. Other states have also created exceptions for employees who are discharged for refusing to violate public policy or who are discharged when an employer refuses to pay for benefits already earned (e.g., sales commissions).

Canadian law has developed a broad-based notice requirement (HRDC 1995b; Levitt 1985). Thus, employers who terminate employees without giving proper notice will generally be required to pay the employee what he or she would have received during the notice period. The notice requirement increases with the length of time on the job.

Canadian employees therefore appear to have a greater level of protection against unjust discharge than U.S. employees. Protection in the U.S. is generally limited to those employees who can demonstrate an employer promise. Protection in Canada is based on employment status, per se. This difference is reflected in the coding of our index. The greatest weight was given to the coverage breadth of the Canadian statutes. Conversely, the situation-specific nature of unjust discharge protection in the U.S. caused us to give it a narrow weight. Thus, all of the Canadian jurisdictions scored a 7, while the U.S. states generally scored between 2 and 3.

Occupational Safety and Health. While it is impossible to analyze detailed regulations for each industry, a comparison of procedures and penalties suggests that U.S. federal safety and health standards are somewhat higher than the standards in the majority of the Canadian provinces. The U.S. index score is 3.13, while the unweighted Canadian average is 2.78. On the other hand, the data indicate that the two of the three largest provinces, Ontario and British Columbia, have fairly high health and safety standards. The standards in Quebec, however, are low relative to other provinces.

The systems in the two countries are distinguishable by their different concepts of monitoring. United States law places a heavy emphasis on governmental monitoring and enforcement through monetary penalties, while Canadian law requires workplace monitoring by workplace health and safety committees.

Advance Notice of Plant Closings/Large Scale Layoffs. The components incorporated into this index include the minimum number of employees necessary to trigger a requirement, the amount of advance notice required, whether notice is given to the affected employees, whether there is a severance pay requirement and whether notice must also be given to the union, if the employees are so represented. The United States has enacted national legislation, the Worker Adjustment and Retraining Notification Act (WARN). Canadian provinces determine the extent of advance notice for firms within their borders.

The basic index averages indicate that the United States has slightly better advance notice provisions than Canada. The basic unweighted scores are 5.03 for the United States and 4.84 for Canada. The basic weighted indices are 5.03 and 4.62 respectively.

These indices, however, are somewhat misleading, since the Canadian scores are influenced by the provinces of Alberta and Prince Edward Island, neither of which have advance notice provisions, If a Canadian index is computed solely for the provinces that have advance notice provisions, the unweighted and weighted averages for Canada increase to 5.72 and 6.34 respectively. This indicates that the Canadian jurisdictions that have enacted advance notice provisions have enacted stricter provisions than WARN in the United States.

ANALYZING THE OVERALL LEVEL OF LABOUR STANDARDS IN CANADA AND THE UNITED STATES

Although summing across the ten labour standards indices provides useful information, it is not an acceptable methodology for performing a cross-standards analysis within and across jurisdictions because the components of each of the standards are not comparable. For this purpose, we developed a procedure based on the ranking of each of the jurisdictions for each of the standards.

In this procedure, each of the 63 (51 in the United States and 12 in Canada) jurisdictions was ranked according to their index scores on each of the standards. For each standard, the jurisdiction with the highest index on that standard was ranked 1 for that standard and the jurisdiction with the lowest index on that standard was ranked 63 on that standard. An adjusted ranking was created by assigning jurisdictions with equal scores a rank equal to the mean of the ranks associated with that score. For example, if two jurisdictions had the highest possible score on a standard, each would be assigned a rank of 1.5 [(1+2)/2].

The ranks for each of the ten standards for each jurisdiction were then summed. Thus the highest possible score (highest labour standards) for a jurisdiction would be 10 (ten standards each ranked first). The lowest possible score for a jurisdiction would be 630 (ten standards each ranked 63rd).

Table 5 presents the results of this ranking analysis. As can be seen, the five jurisdictions with the highest rankings and six of the ten jurisdictions with the highest rankings are Canadian. These jurisdictions include two of the three largest provinces: Ontario and British Columbia. Quebec, the second largest province, ranks fourteenth. The average ranking for the Canadian jurisdictions is 14.92, while the average ranking for the United States jurisdictions is 36.02. The average sum of the adjusted rankings is 252.04 for Canada and 335.99 for the United States.

The provincial autonomy in Canada, however, in contrast to Federal supremacy in the United States also results in a range of adjusted rankings sums that is greater in Canada than in the U.S. Thus, in Canada, the range of the sums of adjusted ranks is 268, while the range of the sums of adjusted ranks in the United States is 182.

In order to determine if the differences between the United States and Canada were significant, Wilcoxon signed-rank tests and Mann-Whitney U tests were conducted for the rankings on each of the standards and sums of each of the standards to determine if these differences could have occurred by change (Mendenhall, Schaeffer and Wackerly 1986). The results, presented in Table 6, show that the Canadian ranks are significantly higher than the United States ranks on six of the ten standards -- paid-time off, unemployment/employment insurance, workers' compensation, collective bargaining, unjust discharge and advance notice [21] -- as well as on the rankings sum. The United States ranks are significantly higher than the Canadian ranks on three standards: minimum wage, overtime and occupational safety and health. There is no significant difference in the two countries in the rankings on the employment equity standards. These results suggest that there is support for the proposition that Canadian labour standards a re, in general, higher than United States labour standards.

SUMMARYAND CONCLUSIONS

This study represents the first step in directly evaluating the differences in labour standards between two countries, in this case Canada and the United States. This was done by creating a means to measure the strength of ten labour standards in the two countries. Each of the standards was evaluated on the basis of an ordinal scale in order to analyze the 10 labour standards across the two countries. Taking the 10 standards as group, assuming that all standards are equally important and that the internal scalings of the standards are identical, the results indicate that, although there are exceptions, Canadian labour standards are higher than U.S. labour standards. This is confirmed by the ranking analysis. Seven of the ten highest ranked jurisdictions are Canadian. Similarly, the average Canadian ranking is 14.92, while the average United States ranking is 36.02. This difference is significant at the .001 level. Thus, a broad-based overview of the labour standards in the two countries suggests that the con ventional wisdom is correct: Canadian labour standards are, indeed, higher than U.S. labour standards.

There are four important caveats that should be noted when assessing this work. First, the creation of the indices depends, to a large extent, on the selection of provisions within each labour standards and the weights given to the various provisions. Excluded provisions or different weights would result in different indices. While we believe our weights are reasonable, other weighting schemes may also be reasonable. [22]

Second, we have defined labour standards quite narrowly, limiting our definition to those standards that appear in both countries and directly effect the employer-employee relationship in both countries. We have excluded such issues as social security, child labour and health insurance, even though, to some extent, these have employment relations implications.

Third, our summing of the indices and the rankings assumes that each of the standards is equally important. It is not likely, however, that all standards will be equally important to all firms and all employees. If a firm produces a labour intensive product and that firm pays relatively low wages, employees and management of those firms may be most concerned about such standards as the minimum wage, overtime and paid-time off. High wage, capital intensive employers may be most concerned about standards that may be perceived as constraining hiring, such as equal employment opportunity/employment equity. High wage, unionized employees may be unconcerned about minimum wage and overtime standards.

These caveats aside, there is a need for a continuing analysis of labour standards for both practical and policy reasons. From a traditional competitive perspective, government imposed labour standards can be seen as a form of intervention in the functioning of the market and the cause of inefficiencies. Inefficient use of scarce resources is seen as a welfare loss for everyone. One important question is whether or not higher labour standards actually do hinder producer competitiveness. In other words, to what extent do labour standards translate into higher production costs that sellers must try to pass onto consumers?

On the other hand, it may be the case that labour standards can be seen as a competitive benefit. By improving the welfare of the labour force as part of a national strategy of encouraging high-wage, high-value added production, the benefits of a competitive economy become widely diffused. Thus, there may not be a trade-off between worker welfare and competitiveness.

Another important practical question with respect to labour standards is the extent to which they are harmonized downward by the pressures of free trade (Gunderson 1998). In this case, at issue is whether countries must lower the cost of their labour to make it more competitive in the global labour market, perhaps regardless of how the country sees its interests (Chaykowski and Giles 1998). Empirically, this has been a difficult question to investigate, in part because analysts have lacked a good set of measures of labour standards that can be incorporated into a comprehensive analysis. We hope that this work will begin to address that gap.

From a policy perspective, the ability to evaluate a jurisdiction's level of labour standards is an important step toward lifting the political debate above the ideological level and forming the basis for informed decision-making. This is particularly important in developed countries where competitive pressures on firms and employees and concerns about the role of government unusually have heightened political pressures to lower the social safety net. A more comprehensive enumeration of labour standards will inform this debate.

- BLOCK, R. N., and K. ROBERTS, School of Labor and Industrial Relations, Michigan State University, East Lansing, Michigan.

- This work was supported by the International Center for Canadian-American Trade (ICCAT), Port Huron, Michigan (through a grant from the United States Department of State), the Canadian Embassy in the United States and the W.E. Upjohn Institute for Employment Research. This article represents the views of the authors and does not necessarily represent those of ICCAT, the United States Department of State, the Canadian Government, or the W.E. Upjohn Institute for Employment Research. The authors wish to thank anonymous referees for helpful comments and Jennifer Evans, Kellie Henehan, Drew Hoffman, Patrick Hughes, Michael Leland, Steven Leonoff, Natasha Kehimkar, Jeonghuyn Lee, Jennifer Lombardo, Jonathan McDermott, Suzanne Monahan, Christine Normoyle, Cynthia Ozeki, Jane Spooner, Chad Taylor and Stephen Whitaker for their research assistance. We also thank Randy Smadella from Statistics Canada for providing employment data.

A book length treatment of this research will appear as Labor Standards in the United States and Canada (title tentative), to be published by the W.E. Upjohn Institute for Employment Research in Kalamazoo, Michigan. At that time, the W.E. Upjohn Institute will make the data used to compute the indices herein publicly available through the World Wide Web.

(1.) Recent examples of this concern include the 1997 political debate in the United States regarding granting the President of the United States "fast track" authority to negotiate trade agreements and the 1999 demonstrations associated with the meeting of the World Trade Organization in Seattle, Washington (Mitchell 1997; "Trade Delegates..." 1999).

(2.) Traditional neoclassical economic trade theory has focussed on relative factor (capital and labour) prices, with little attention to the distributional effects of trade on workers (Jepma, Jaeger and Kamphius 1996). Indeed, labour issues are often seen as impairing the attempts of policymakers attempting to encourage free trade (see, e.g., Cooper 1997; Mitchell 1997). Modern frameworks which analyze trade, however, address the behaviour of business, government and labour in the trading system (Jepma, Jaeger and Kamphius 1996). In its December 1996 Ministerial Declaration, the World Trade Organization called for the "observance of internationally recognized core labor standards" (World Trade Organization 1996). For a discussion of the effects of trade on one aspect of labour welfare, income distribution, see, for Burtless (1995). For a more general discussion of trade theory and labour standards, see the methodology section below.

(3.) There has been some work on specific issues that might be considered labour standards. For comparative work in collective bargaining, see Weiler (1983), Adams (1993) and Block (1994, 1996). For comparative studies in workers' compensation, see Burton (1989), Roberts and Madden (1992), and Hyatt and Kralj (1992). For a comparative study in unjust discharge, see Jain (1992). For an overview of social contracts, see Card and Freeman (1994).

(4.) Not all work finds that U.S. labour costs are lower than those in Canada. One study (Roberts and Smith 1992) found that U.S. and Canadian labour costs were quite comparable. Expressed in U.S. dollars, in 1993, average hourly compensation for production workers in manufacturing stood at $16.79 in the United States and $16.36 in Canada, further suggesting comparability between the two countries (U.S. Department of Labor 1994). To the extent that labour standards are reflected in hourly compensation, this research should cause one to question the proposition that Canadian firms are at a disadvantage relative to their counterparts in the United States.

(5.) See Compa (1993), Charnovitz (1987), and Piore (1990) for examples of other lists of labour standards.

(6.) Restrictions on child labour were excluded because we view them as a blend of social welfare and labour standards. Our focus was limited to those labour market laws that are clearly labour standards. For a discussion of the labour market implications of child labour, see Basu and Van (1998).

(7.) Data for the coding the provisions were obtained from the sources listed below:

Minimum Wage. Canada: HRDC (1995c); United States: H.R. 3448, Small Business Job Protection Act of 1996.

Overtime. Canada: Federal: Canada Labour Code and Labour Standards Regulations;

Alberta: Employment Standards Code and Reg. 81/81; B.C. Employments Standards Act and Regulation; Manitoba Employment Standards Act; New Brunswick Minimum Wage Regulation; Newfoundland, Labour Standards Act; Newfoundland Labour Standards Act and Regulations; Northwest Territories Employment Standards Act; Nova Scotia Labour Standards Code and Remembrance Day Act; Ontario Employment Standards Act; Prince Edward Island Employment Standards Act; Quebec Act Respecting Labour Standards and National Holiday and Labour Standards Act and Regulations; Saskatchewan Labour Standards Act; Yukon Employment Standards Act regulations. United States: Bureau of National Affairs (1995a, 1995b).

Paid-Time Off. Canada: Federal: Canada Labour Code and Labour Standards Regulations; Alberta: Employment Standards Code and Reg. 81/81; B.C. Employments Standards Act and Regulation; Manitoba Vacations with Pay Act, Employment Standards Act, and Remembrance Day Act; New Brunswick Employment Standards Act; Newfoundland Labour Standards Act; Northwest Territories Labour Standards Act; Nova Scotia Labour Standards Code and Remembrance Day Act; Ontario Employment Standards Act; Prince Edward Island Employment Standards Act; Quebec Act Respecting Labour Standards and National Holiday and Labour Standards Act and Regulations; Saskatchewan Labour Standards Act; Yukon Employment Standards Act. United States: Bureau of National Affairs (1995b).

Unemployment/Employment Insurance. Canada: Government of Canada (1995), HRDC (1997) and various HRDC sources on employment insurance; United States: Bureau of National Affairs (1995b).

Workers' Compensation. Canada: U.S. Chamber of Commerce (1996). United States: Burton and Schmidle (1996).

Collective Bargaining. Canada: Adams (1997); United States: Hardin (1990). Nondiscrimination-Equal Employment Opportunity/Employment Equity. Canada: Kelly (1991), Aeberhard-Hodges (1996); United States: Wolkinson and Block (1996).

Unjust Dismissal. Canada: Levitt (1985), HRDC (1995b); United States: Bureau of National Affairs (1994).

Occupational Safety and Health. Canada: Commerce Clearinghouse (various years); United States: Wolkinson and Block (1996).

Advance Notice of Plant Closings/Large-Scale Layoffs. Canada: HRDC (1995a) and links to provincial sites, telephone calls to provincial ministries of labour; United States: Worker Adjustment and Retraining Notification Act.

The data obtained through these sources were further augmented and updated on the basis of information provided on the websites of the various Canadian jurisdictions.

(8.) The full spreadsheets showing each statutory provision within each jurisdiction are available upon request.

(9.) An attempt to compare agency budgets per covered employer or employer as a measure of aggressiveness of enforcement was abandoned due to differing functions and missions of agencies between the two countries and, on occasion, between jurisdictions within the two countries.

(10.) We do not include appeals rights/judicial review as a consideration in unemployment insurance (United States) I employment insurance (Canada) because these cases are highly individualistic and have no effect on the general standards. An additional element was included in the workers' compensation enforcement measures: the scope of appeal within the workers' compensation adjudicative system. Typically, jurisdictions allow either a narrow review based solely on a point of law or a more complete review that permits a full review of the record, sometime to the point of a de novo appeal.

(11.) For a somewhat comparable process of coding legislation, albeit only for public sector collective bargaining laws, see, Currie and McConnell (1991) and Gunderson, Hebdon and Hyatt (1996). For other analyses of labour standards across countries, see Rodrick (1994) who measured a country's labour standards as the total number of International Labour Organization (ILO) conventions and basic rights ILO conventions ratified by the country, indicators of political and civil rights in the country, indicators of the extent of child labour legislation annual enforcement, statutory hours of work in manufacturing or construction, days of annual leave in manufacturing and the percentage of the labour force unionized; Aggarwal (1995a, 1995b) who reports export and investment data for countries and sectors considered to have varying levels of labour standards and OECD (1996) country's ratification of ILO conventions.

(12.) To some extent, these indices are a function of the weights given to each relevant provision. While the weights are open to debate, we believe that the weighting scheme developed is reasonable, although other weighting schemes are likely equally to be reasonable.

(13.) There are no occupational or industrial exclusions from judicially determined exceptions to employment-at-will in the United States.

(14.) The only departure from this assumption was for British Columbia, where managers are exempt.

(15.) A full discussion of the procedure for estimating coverage, including the general form equations, is available from the authors upon request.

(16.) Total employment for the Yukon and the Northwest Territories was not included in the data set. The employment totals for these two jurisdictions were taken from the following internet addresses: http://www.yukonweb.com/govermnent/facts.htmld/#earnings and http://www.stats.gov.nt.ca/Bureau/Statlnfof/LabourForce/_LfsData.html .

(17.) The data from which these indices were derived are available from the authors upon request.

(18.) Employment data for the United States were based on an average of the percentage of total employment in each state in May 1995 and February 1999 and were derived from U.S. Bureau of Labor Statistics, "Employees on Nonfarm Payrolls, by State and Selected Industry Division," at http://stats.bls.gov/news.release/laus.t05.htm. Employment data for Canada was an average of the percentage of employment in each province/territory in 1995 and 1999 and were obtained from Statistics Canada, "Labour Force, Employed and Unemployed, Numbers and Rates," at http://www.statcan.ca/english/Pgdb/People/Labour. The employment-weighted average does not distinguish between employees in the federal jurisdiction and employees in the provincial/territorial jurisdiction. To separate the federal jurisdiction would require us to subtract from each province's employment estimate the number of employed persons in the province who are employed in an industry in the federal jurisdiction. At this stage, our data do not permit us to make such a calculation.

(19.) For the purposes of this paper, we assume that there is no shifting of the incidence of the unemployment insurance premiums from the employer to the employee in the United States. Put differently, we do not take account of the possible economic effects and focus solely on the legal incidence.

(20.) The data used to construct the index are taken from the 1997 report (U.S. Department of Labor 1997) and a reprint of their 1995 report in Burton and Schmiddle (1996). Compliance among Canadian provinces was derived from two primary sources: Burton and Scbmiddle (1996) and U.S. Chamber of Commerce (1996).

(21.) The advance notice index is the only one of the standards for which the inference from the statistical tests differs substantially from the inference based on the visual inspection of the index means. This results from the fact that the majority of Canadian jurisdictions that have enacted advance notice legislation have enacted fairly strong provisions and are ranked quite high. These more than offset the low ranks of the few Canadian jurisdictions that have not enacted advance notice provisions.

(22.) The data will be made available through the W.E. Upjohn Institute for Employment Research to any researcher who wishes to recompute the indices. For a colloquy on the reasonableness of a coding scheme for public sector collective bargaining laws in Canada, see Gunderson, Hebdon and Hyatt (1996) and Currie and McConnell (1996).

* REFERENCES

ADAMS, George W. 1995. "Supplement: Review of Recent Amendments to Ontario's Labour Legislation." Canadian Labour Law. 2nd ed. Aurora, Ont.: Canada Law Book, November.

ADAMS, George W. 1997. Canadian Labour Law. 2nd ed. Aurora, Ont.: Canada Law Book, May.

ADAMS, Roy J. 1993. "The Role of the State in Industrial Relations." Research Frontiers in Industrial Relations. D. Lewin, O. Mitchell and P. Sherer, eds. Madison, Wisc.: Industrial Relations Research Association Series.

ADDISON, John, Douglas Fox and Christopher RUHM. 1995. "Trade and Displacement in Manufacturing." Monthly Labor Review, Vol. 188, No. 6, 58-66.

AEBERHARD-HODGES, Jane. 1996. "Sexual Harassment in Employment: Recent Judicial and Arbitral Trends." International Labour Review, Vol. 135, No. 5, 499-531.

AGGARWAL, Mita. 1995a. "International Trade, Labor Standards, and Labor Market Conditions: An Evaluation of the Linkages." U.S. International Trade Commission, Office of Economics Working Paper 95-06-C, June.

AGGARWAL, Mita. 1995b. "International Trade and the Role of Labor Standards." International Economic Review, August, 22-30.

BASU, Kaushik and Pahm HOANG VAN. 1998. "The Economics of Child Labor." The American Economic Review, Vol. 88, No. 3, 412-427.

BLANK, Rebecca. 1994. "Does a Larger Social Safety Net Mean Less Economic Flexibility?" Working under Different Rules. R. Freeman, ed. New York: Russell Sage Foundation, 157-187.

BLOCK, Richard N. 1978a. "The Impact of Seniority Provisions on the Manufacturing Quit Rate." Industrial and Labor Relations Review, Vol. 31, No. 4, 474-488.

BLOCK, Richard N. 1978b. "Job Changing and Negotiated Nonwage Provisions." Industrial Relations, Vol. 17, No. 3, 296-307.

BLOCK, Richard N. 1992. "The Legal and Institutional Framework for Employment Security in the United States: An Overview." Employment Security and Labor Market Flexibility: An International Perspective. K. Koshiro, ed. Detroit, Mich.: Wayne State University Press, 127-148.

BLOCK, Richard N. 1994. "Reforming U.S. Labor Law and Collective Bargaining: Some Proposals Based on the Canadian Industrial Relations System." Restoring the Promise of American Labor Law. S. Friedman, R. Hurd, R. Oswald and R. Seeber, eds. Ithaca, NY: ILR Press, 250-259.

BLOCK, Richard N. 1996. "NAFTA, Collective Bargaining and Employment Adjustments in the United States and Canada." Policy Choices: Free Trade among NAFTA Nations. K. Roberts and M. Wilson, eds. East Lansing, Mich.: Michigan State University Press, 217-234.

BLOCK, Richard N. 1997a. "Rethinking the National Labor Relations Act and Zero-Sum Labor Law: An Industrial Relations View." Berkeley Journal of Employment and Labor Law, Vol. 18, No. 1, 30-55.

BLOCK, Richard N. 1997b. "Collective Bargaining and the Courts: An Empirical Analysis of Judicial Review of National Labor Relations Board Decisions and Arbitration Awards." Paper presented at the Twenty-Fifth Annual Conference of the National Center for the Study of Collective Bargaining in Higher Education and the Professions, School of Public Affairs, Baruch College, CUNY, New York City, April 14-15 (publication forthcoming).

BRUCE, Peter D. 1990. "The Processing of Unfair Labor Practice Cases in the United States and Canada." Relations industrielles/Industrial Relations, Vol. 45, 481-5 11.

BRUDNEY, James J. 1996. "A Famous Victory: Collective Bargaining Protections and the Statutory Aging Process." North Carolina Law Review, Vol. 74, No. 2, 938-1036.

BUREAU OF NATIONAL AFFAIRS. 1994. Individual Employment Rights Reference Manual. 505, March. Washington D.C.: Bureau of National Affairs.

BUREAU OF NATIONAL AFFAIRS. 1995a. Fair Labor Standards Act. Labor Relations Reporter: Labor Relations Expediter. Washington D.C.: Bureau of National Affairs.

BUREAU OF NATIONAL AFFAIRS. 1995b. State Laws. State Policy and Practice Series, part 2. Washington D.C.: Bureau of National Affairs.

BUREAU OF NATIONAL AFFAIRS. 1995c. Fair Labor Standards Act. Labor Relations Reporter: Wage and Hour Manual. Washington D.C.: Bureau of National Affairs.

BURTLESS, Gary. 1995. "International Trade and the Rise in Earnings Inequality." Journal of Economic Literature, Vol. 33, No. 2, 800-816.

BURTON, John F., Jr. 1989. "Workers' Compensation in Canada and the United States: Decentralized Approaches and Divergent Outcomes." John Burton's Workers' Compensation Monitor, January, 7-12.

BURTON, John F. Jr. and Timothy SCHMIDLE. 1996. 1996 Workers' Compensation Yearbook. Horsham, PA: LRP Publications.

CARD, David and Richard FREEMAN. 1994. "Small Differences that Matter: Canada vs. the United States." Working Under Different Rules. R. Freeman, ed. New York: Russell Sage Foundation, 189-221.

CHAISON, Gary N. and Joseph B. ROSE. 1994. "The Canadian Perspective on Workers' Rights to Form a Union and Bargain Collectively." Restoring the Promise of American Labor Law. S. Friedman, R. Hurd, R. Oswald and R. Seeber, eds. Ithaca, NY: ILR Press, 241-249.

CHAYKOWSKI, Richard and Anthony GILES. 1998. "Globalization, Work, and Industrial Relations." Relations industrielles/Industrial Relations, Vol. 53, No. 1, 3-23.

CLARK, BOARDMAN, CALLAGEN. 1995. Employment Coordinator. Washington, DC.

COMMERCE CLEARINGHOUSE. various years. Canadian Employment Safety and Health Guide (each province).

COMPA, Lance. 1993. "Labor Rights and Labor Standards in International Trade." Law & Policy in International Business, Vol. 25, No. 1, 165-191.

COOPER, Helene. 1997. "Clinton Seeks More Leeway in Trade Pacts." Wall Street Journal, April 1, A3.

CROWLEY, Donald W. 1987. "Judicial Review of Administrative Agencies." Western Political Quarterly, Vol. 40, No. 2, 265-283.

CURRIE, Janet and Sheena MCCONNELL. 1991. "The Effect of the Legal Structure of Collective Bargaining on Dispute Costs and Wages." American Economic Review, Vol. 81, No. 4, 693-718.

CURRIE, Janet and Sheena MCCONNELL. 1996. "Collective Bargaining in the Public Sector: Reply." American Economic Review, Vol. 81, No. 4, 327-328.

DE WAART, Paul. 1996. "Minimum Labour Standards in International Trade from a Legal Perspective." Challenges to the New World Trade Organization. Pitou van Dijck and Gerrit Faber, eds. The Hague/London/Boston: Kluwer Law International, 245-264.

DOERN, G. Bruce and Brian W. TOMLIN. 1991. Faith and Fear: The Free Trade Story. Toronto: Stoddart Publishing Co.

DORMAN, Peter. 1995. "Policies to Promote International Labor Rights: An Analytical Review." Paper prepared for the Bureau of International Labor Affairs, U.S. Department of Labor.

GOVERNMENT OF CANADA. 1988. Canada Labour Code, R.S.C. 1985, c. L-2.

GOVERNMENT OF CANADA. 1993. Canada Labour Code, R.S.C. 1985, c. 27.

GOVERNMENT OF CANADA. 1995. A 21st Century Employment System for Canada: Employment Insurance Legislation Highlights. Ottawa.

GREENHOUSE, Steven. 1997. "Nike Shoe Plant in Vietnam Is Called Unsafe for Workers." The New York Times, November 8, Sec. A., 1.

GUNDERSON, Morley. 1998. "Harmonization of Labour Policies Under Trade Liberalization." Relations industrielles/Industrial Relations, Vol. 53, No. 1, 24-54.

GUNDERSON, Morley, Robert HEBDON and Douglas HYATT. 1996. "Collective Bargaining in the Public Sector: Comment." American Economic Review, Vol. 86, No. 1, 315-326.

HARDIN, Patrick, et al. 1990. The Developing Law. 3rd ed. Washington, D.C.: Bureau of National Affairs.

HUMAN RESOURCES DEVELOPMENT CANADA (HRDC). 1995 a. Employment Standards Legislation in Canada, 1995-96 edition. Ottawa, Canada.

HUMAN RESOURCES DEVELOPMENT CANADA (HRDC). 1995b. Industrial Relations Legislation in Canada. Ottawa, Canada.

HUMAN RESOURCES DEVELOPMENT CANADA (HRDC). 1995c. Information on Labour Standards. Ottawa, Canada.

HUMAN RESOURCES DEVELOPMENT CANADA (HRDC). 1997. The New Employment Insurance System. http://www.hrdc-drhc.gc.ca/hrdc/ei/ina068_e.html.

HYATT, Doug and Boris KRALJ. 1992. "Workers' Compensation Costs and Competitiveness: Issues and Inter-Jurisdictional Comparisons." Women and Industrial Relations. D. Carter, ed. Queen's University, Kingston, Ont., 421-430.

INTERNATIONAL LABOR ORGANIZATION (ILO). 1995. Lists of Ratifications by Convention and by Country as of 3l December, 1994. International Labour Conference, 82nd Session, Report III, Part 5.

JAIN, Hem. 1992. "Wrongful Dismissals in Non-Union Companies in North America." Women and Industrial Relations. D. Carter, ed. Queen's University, Kingston, Ont., 195-202.

JEPMA, Catrinus J., Henk JAGER and Elise KAMPHIUS. 1996. Introduction to International Economics. London and New York: Longman/Heerlen, Neth.: Open University of the Netherlands.

KELLY, John G. 1991. Human Resource Management and the Human Rights Process. 2nd ed. Don Mills, Ont.: CCH Canada Ltd.

KOCHAN, Thomas A. and Richard N. BLOCK. 1977. "An Interindustry Analysis of Bargaining Outcomes: Preliminary Evidence from Two-Digit Industries." Quarterly Journal of Economics, Vol. 91, No. 3, 431-452.

KUMAR, Pradeep. 1991. Canadian Labour Relations: An Information Manual. Kingston, Ont: Industrial Relations Centre, Queens University.

LEE, Joseph S., ed. 1996. Labor Standards and Economic Development. Taipei, Taiwan: Chung-Hua Institution for Economic Research.

LEVITT, Howard. 1985. The Law of Dismissal in Canada. Aurora, Ont.: Canada Law Books.

LIPSET, Seymour Martin. 1989. Continental Divide: The Values and Institutions of the United States and Canada. Toronto and Washington, D.C.: C.D. Howe Institute and National Planning Association.

LYON, Peyton. 1987. "CUFTA and Canadian Independence." Canadian-American Free Trade: Historical, Political, and Economic Dimensions. A.R. Riggs and Tom Velk, eds. Halifax, N.S: The Institute for Research on Public Policy.

MAHANT, Edelgard E. 1993. Free Trade in American-Canadian Relations. Malabar, Fla.: Krieger Publishing Co.

MARTIN, Richard. 1991. "Evaluating Free Trade: A Perspective from the CLC." The Economic Impact and Implications of the Canada-U.S. Free Trade Agreement. Fakhari Siddiqui, ed. Lewiston, N.Y. and Queenston, Ont.: The Edwin Mellon Press, 25-36.

MENDENHALL, William, Robert SHAEFFER, and Dennis D. WACKERLY. 1986. Mathematical Statistics With Applications. 3rd ed. Boston, Mass.: Duxbury.

MITCHELL, Alison. 1997. "Clinton Clears First Hurdle on Fast Track Trade Legislation." The New York Times, November 4.

MULLAN, David. 1993. "Judicial Deference to Executive Decision-Making: Evolving Concepts of Responsibility." Queen's Law Journal, Vol. 19, No. 1, 137-178.

NATIONAL COMMISSION ON STATE WORKMEN'S COMPENSATION LAWS. 1972. Report of the National Commission on State Workmen's Compensation Laws. U.S. Government Printing Office.

NORTH AMERICAN AGREEMENT ON LABOR COOPERATION. 1993. Between the Government of the United States of America, the Government of Canada, and the Government of the United Mexican States, September 13 at: gopher://cyfer.esueda.gov70/oo/ace/policy/nafta/labor-co.txt.

NOVAK, Rita M. and Douglas K. SOMERLOT. 1990. Delay on Appeal: A Process for Identifying Causes and Cures, Chicago, Ill.: American Bar Association.

ORGANIZATION FOR ECONOMIC COOPERATION AN DEVELOPMENT (OECD). 1996. Trade, Employment and Labour Standards: A Study of Core Workers' Rights and International Trade. Paris: OECD.

PIORE, Michael. 1990. "Labor Standards and Business Strategies." Labor Standards and Development in the Global Economy. S. Herzenberg and J. Perez-Lopez, eds. Washington, D.C.: U.S. Department of Labour, Bureau of International Labor Affairs, 35-49.

ROBERTS, Karen and Michael MADDEN. 1992. "Workers' Compensation as an Economic Development Issue." Women and Industrial Relations. D. Carter, ed. Queen's University, Kingston, Ont., 409-420.

ROBERTS, Karen and Philip SMITH 1992. "The Effect of Labor Cost Differences on the Location of Economic Activity under the U.S.-Canada Free Trade Agreement." Economic Development Quarterly, Vol. 61, January, 52-63.

RODRICK, Dani. 1994. "Labor Standards in International Trade: Do They Matter and What Do We Do About Them?" Overseas Development Council, Policy Essay No. 20, 35-79.

RUGMAN, Alan. 1991. "Adjustments by Multinational Firms to Free Trade." Economic Opportunities in Freer U.S. Trade with Canada. F. Menz and S. Stevens, eds. Albany, NY: State University of New York Press, 53-66.

SCHNEID, Thomas. 1992. The Americans with Disabilities Act. New York: Van Nostrand Reinhold.

SMITH, Murray. 1988. "The Free Trade Agreement in Context." The Canada-United States Free Trade Agreement: The Global Impact. J. Schott and M. Smith, eds. Washington, D.C.: Institute for International Economics, 37-64.

STATISTICS CANADA. 1998. Average Weekly Hours, For Employees Paid by the Hour. 72F0002-XDE at http://www.statcan.ca:80/ english/ Pgdb/ People/ labour.htm.

STEVENSON, Richard W. 1997. "Executives Say Trade Bill Defeat Will Hurt Economy." The New York Times, November 12.

"Trade Delegates Close in on a Deal to Start Future Trade Negotiations." 1999. Wall Street Journal Interactive Edition, December 3.

U.S. CHAMBER OF COMMERCE. 1995. Analysis of Workers' Compensation Laws. Washington, D.C.

U.S. CHAMBER OF COMMERCE. 1996. 1996 Analysis of Workers' Compensation Laws. Washington, D.C.

U.S. DEPARTMENT OF LABOR. 1994. "International Comparisons of Manufacturing Hourly Compensation Costs, 1993." News: United States Department of Labor, Bureau of Labor Statistics, USDL-94-261, May 24.

U.S. DEPARTMENT OF LABOR. 1997. State Workers' Compensation Laws. Washington, D.C.: Employment and Standards Administration, January.

WEILER, Paul C. 1983. "Promises to Keep: Securing Workers' Rights to Self-Organization Under the NLRA." Harvard Law Review, Vol. 96, 1769-1827.

WHITE, Randall. 1988. Fur Trade to Free Trade: Putting the Canada-U.S. Free Trade Agreement in Historical Context. Toronto and Oxford: Dundern Press.

WOLKINSON, Benjamin and Richard N. BLOCK. 1996. Employment Discrimination: The Workplace Rights of Employees and Employers. Cambridge, Mass., USA and Oxford, UK: Blackwell Publishers.

World Trade Organization. 1996. "Singapore Ministerial Declaration." 13 December 1996, World Trade Organization Ministerial Conference, Singapore, 9-13 December at http://www.wto.org/wto/Whats_new/wtodec.htm.

R[acute{E}]SUM[acute{E}]

Les normes du travail aux [acute{E}]tats-Unis et au Canada

La validit[acute{e}] de la pr[acute{e}]somption g[acute{e}]n[acute{e}]ralis[acute{e}]e [grave{a}] l'effet que les normes du travail sont moins g[acute{e}]n[acute{e}]reuses aux [acute{E}]tats-Unis qu'au Canada n'a jamais [acute{e}]t[acute{e}] v[acute{e}]rifi[acute{e}]e et les diff[acute{e}]rences ce chapitre entre ces deux pays n'ont pas non plus fait l'objet d'une analyse approfondie. C'est ce que nous tentons de faire ici.

Comme premi[grave{e}]re [acute{e}]tape. il nous est apparu n[acute{e}]cessaire de b[hat{a}]tir une d[acute{e}]finition de [ll] normes du travail [gg] qui puisse [hat{e}]tre applicable aux deux pays. Cette d[acute{e}]finition est: toute proc[acute{e}]dure, condition d'emploi ou exigence impos[acute{e}]es a un employeur par le gouvernment qui a pour objectif la protection des employ[acute{e}]s contre un traitement au travail que la soci[acute{e}]t[acute{e}] consid[grave{e}]re injuste et in[acute{e}]quitable.

[grave{A}] partir de cette d[acute{e}]finition, nous avons [acute{e}]tabli que les normes du travail peuvent [hat{e}]tre class[acute{e}]es en deux cat[acute{e}]gories (1) ces normes qui exigent des paiements mon[acute{e}]taires par l'employeur soit aux travailleurs ou [grave{a}] une agence gouvemementale; (2) ces nonnes qui imposent des contraintes [grave{a}] l'employeur eu [acute{e}]gard a ses actions vis-[grave{a}]-vis les travailleurs.

Les normes de la premi[grave{e}]re cat[acute{e}]gorie ici analys[acute{e}]es sont celles de salaire minimum, de surtemps, de cong[acute{e}]s, d'assurance-emploi et d'accidents du travail. Parmi celles de la deuxi[grave{e}]me cat[acute{e}]gorie, nous avons retenu celles de Ia n[acute{e}]gociation collective, de la non-discrimination, du cong[acute{e}]diement injuste, de la sant[acute{e}] et de la s[acute{e}]curit[acute{e}] du travail et du pr[acute{e}]avis de fermeture d'entreprise ou de licenciements massifs. Toutes ces normes correspondent [grave{a}] notre d[acute{e}]finition et impliquent toute intervention ou r[acute{e}]glementation gouvernementale au lieu de travail.

Notre m[acute{e}]thodologie de comparaison a trois composantes: (1) une analyse du contenu de chacune des normes incluant les dispositions r[acute{e}]glementaires et la nature de leur application; (2) la construction d'un indice de la force d'une norme dans une juridiction en utilisant une approche pond[acute{e}]r[acute{e}]e pour les diff[acute{e}]rentes positions r[acute{e}]glementaires pour chacune de ces normes; (3) la d[acute{e}]flation des normes pond[acute{e}]r[acute{e}]es par une estimation du pourcentage de la main-d'oeuvre couverte par chaque norme.

Nous avons [acute{e}]tabli un indice de base pour chaque norme dans chacune des juridictions, soit les [acute{e}]tats pour les [acute{E}]tats-Unis, les provinces et les territoires pour le Canada, indice qui fournit une mesure de chaque norme telle qu'elle affecte le travailleur typique. Pour effectuer des comparaisons entre les pays, nous avons imagin[acute{e}] d'une part une moyenne non pond[acute{e}]r[acute{e}]e pour chacune des normes pour chacun des deux pays et, d'autre part, une moyenne pond[acute{e}]r[acute{e}]e pour la part d'emploi de chaque juridiction dans l'emploi total de son pays.

Nous [acute{e}]tablissons une mesure d'ensemble du bien-[hat{e}]tre des employ[acute{e}]s associ[acute{e}]e aux normes du travail en d[acute{e}]flationnant chaque norme applicable par le pourcentage de la main-d'oeuvre couverte. Comme tel, cela fournit plus une mesure g[acute{e}]n[acute{e}]rale, de niveau soci[acute{e}]tal plut[hat{o}]t qu'un indice de base, mesure que nous appelons l'indice d[acute{e}]flationn[acute{e}].

Nous avons donn[acute{e}] des valeurs pour chacune des dispositions r[acute{e}]glementaires pertinentes ou pour le m[acute{e}]canisme d'application. Les valeurs vont de 0 (absence de disposition) [grave{a}] 10 (dispositions les meilleures). Les dispositions de force interm[acute{e}]diaire appelaient des valeurs interm[acute{e}]diaires selon le nombre de cat[acute{e}]gories possibles dans cette disposition.

C'est en utilisant cette [acute{e}]chelle que nous avons analys[acute{e}] les dix normes du travail pour les deux pays. En consid[acute{e}]rant ces dix normes comme un groupe et en faisant l'hypoth[grave{e}]se qu'elles sont toutes [acute{e}]galement importantes et que l'[acute{e}]chelonnage interne de ces normes est identique, les r[acute{e}]sultats d[acute{e}] montrent, s'il yades exceptions, que les normes canadiennes sont sup[acute{e}]rieures aux normes am[acute{e}]ricains.

Ceci est confirm[acute{e}] par une analyse de rangement. Six des dix juridictions classifi[acute{e}]es comme les plus hautes sont canadiennes. De plus, le rangement moyen canadien est de 14,92 alors qu'il est de 36,02 pour les [acute{E}]tats-Unis. Cette diff[acute{e}]rence est significative [grave{a}] un niveau tie ,001. 1. Notre [acute{e}]tude permet de dire que la croyance populaire est correcte: les normes canadiennes du travail sont sup[acute{e}]rieures aux normes am[acute{e}]ricaines.