Insufficient demand

It is still widely believed that insufficient aggregate demand is largely responsible for high unemployment levels. In reality, however, the magnitude and long upward trend of the unemployment rate cannot be explained by inadequate spending. [Herb Grubel, "Labour Market Rigidities and Canada's High Unemployment Rates", paper presented at The Fraser Institute Conference "Right-to-Work Laws: The Global Evidence of Their Impact in Reducing Unemployment," Calgary, May 16, 1997, for a full exposition of why inappropriate aggregate demand policies are not a significant cause of unemployment] During periods of recession, the cyclical reduction in aggregate demand undoubtedly results in a temporary increase in the rate of unemployment. But in Canada's case, successive economic recoveries have not seen the unemployment rate return to its pre-recession levels. Instead, the rate has moved upwards and today remains high, despite a relatively strong economic performance in recent years.

New technologies

Another mistaken but widely held belief is that the current high rate of unemployment is due to the introduction of new technologies and labour-saving production techniques. This reasoning also does not explain the problem in Canada, given that in the past, technological change has not led to high and sustained unemployment rates and, more importantly, the same new technologies have been introduced in the US to an even greater extent where currently the unemployment rate is almost half of Canada's.[Grubel, "Labour Market Rigidities," p. 1]

Payroll taxes

More recently, payroll taxes have come under scrutiny as a primary cause of rising unemployment because they add to an employer's total wage bill. The belief is that payroll taxes increase the cost of hiring an employee which reduces a firm's demand for labour. The substantial rise in payroll taxes that has occurred over the last two decades probably has reduced employment to some extent at different times. [Livio De Matteo and Michael Shannon, "Payroll Taxation in Canada: An Overview," Canadian Business Economics, vol. 3, no. 4, Summer 1995, p. 5-22] In the longer run, however, companies respond to the imposed costs by limiting future pay increases, and thus over time adjust to the new costs. Although payroll taxes may have a short-term negative effect on employment rates, they cannot account for the sustained, high rates of unemployment experienced in the last decade.

Government downsizing

A final misperception would have government downsizing and public sector layoffs responsible for the current high unemployment rates. Again, because these layoffs have occurred in significant numbers only recently, they cannot account for the fact that unemployment rates have been rising since the early 1970s. Moreover, the unemployment associated with government downsizing should also only be temporary as these workers can find employment in the private sector. All of the above factors are associated mostly with temporary increases in unemployment rates. Although they may contribute somewhat to Canada's current high rates of joblessness, they cannot account for either the magnitude or long term upward trend in Canada's unemployment rate. Furthermore, they cannot explain why Canada has unemployment rates substantially higher than comparable rates in other OECD countries.

While the "usual suspects" cannot be found responsible for our unemployment rate, there is a growing consensus among economists that the most significant cause of Canada's unemployment is excessive government restrictions on labour market operations. Ironically, it is becoming increasingly evident that policies intended to protect individual workers, such as employment insurance, minimum wage laws, employment standards regulations, welfare benefits, and regulations governing unions, actually impose large costs on society in the form of excessive unemployment. [See Walter Block and Michael Walker, eds., Discrimination, Affirmative Action and Equal Opportunity, The Fraser Institute, 1982; Herb Grubel and Michael Walker, eds., Unemployment Insurance: Global Evidence of its Effects on Unemployment, The Fraser Institute, 1978; Ernst and Young, B.C. Minimum Wage Study: Canadian Restaurant and Foodservices Association Final Report, 1995; Fazil Mihlar, "Reality Check: Minimum Wage Laws do Kill Jobs," Fraser Forum, June 1997; OECD, The OECD Jobs Study: Evidence and Explanations-Parts I and II.]The focus of this study, therefore, is on labour legislation governing unions, and the related economic impact of unions.

To provide a comprehensive explanation of the unemployment problem and a partial solution, this study discusses a combination of theoretical background and empirical evidence. The first section provides background information on labour laws in Canada and summarizes the proposed changes to the Canada Labour Code. The next section discusses the requisite theoretical material on unemployment, the economic impact of unions, and Canada's unemployment problem. The third section documents the necessity for, and process of, labour market reform in both Britain and New Zealand. Finally, the study concludes with policy recommendations for the Canada Labour Code, based largely on the successful reforms in Britain and New Zealand.


Bill C-66: Strengthening Unions?

The premise of this paper is straightforward. Governments can affect long term unemployment rates, not with fiscal or monetary initiatives per se, but through legislation and institutions governing labour relations.[There are limits to what fiscal and monetary policy can achieve (especially in the longer term). For a discussion of the failure of Keynesian macroeconomic models (used by governments), to account for supply side issues, see Peter S. Spiro, "The Effects of Fiscal Policy on Canadian Economic Growth and Employment", Ontario Ministry of Finance, Working Paper, 1996] In countries such as Britain and New Zealand, legislation that afforded unions excessive privileges and powers culminated in disastrous economic consequences, including high rates of unemployment and slow economic growth. As is discussed below, British labour regulations provided unions with so much protection that they wielded enough power to effectively shut the economy down during the 1978-79 "Winter of Discontent." In New Zealand, legislation was so skewed that unions could be formed by a few individuals and contracts negotiated with only a select number of employers. Even more extreme was that under the law the contracts (negotiated only with "chosen" employers) applied to all companies with workers in the same craft or occupation. In both countries, labour legislation contributed to a severe misallocation of labour resources. Ultimately, in order to combat high rates of unemployment and poor economic growth, both countries introduced reforms that dramatically liberalized labour markets. As a result, Britain is now enjoying one of the lower unemployment rates among European countries, and New Zealand is basking in a rate approaching that of the United States.

Under Canadian legislation, union privileges and powers are not nearly as unbalanced as they were in Britain and New Zealand. However, unions are granted a number of legally sanctioned rights designed to protect them, and to facilitate the certification process. Most significant for the union movement is the principle of exclusive representation. Although the principle varies somewhat across jurisdictions, unions in Canada have the right to represent all workers in a bargaining unit. This legislated right provides unions with a significant amount of security at the expense of an individual's right to negotiate his or her employment arrangements.[Dennis Maki, "Canada's Labour Laws: Where Do We Stand?" in Fazil Mihlar, ed., Unions and Right to Work Laws: The Global Evidence of Their Impact on Employment, The Fraser Institute, 1997]

The strongest form of union security is the "closed shop," where the company and union both agree that all workers must be union members prior to becoming employed by the company. The somewhat less restrictive "union shop" permits management to hire anyone they see fit, but still requires all workers to join the union and be bound by the work rules it imposes.[Sometimes exemptions from mandatory membership may apply to workers such as those who were employed prior to certification. See "Canada's Labour Laws: Where Do We Stand?"]

The "check-off" system requires employers to collect union dues from employees on behalf of the union .[" Canada's Labour Laws: Where Do We Stand?"] The degree of union security also varies under this collect-and-remit principle. For example, provisions may be such that employers collect dues only after workers voluntarily authorize them to do so. A stronger variant requires employers to collect dues from all union members without regard to authorization. The strongest version of the check-off system, most commonly referred to as the Rand Formula (or "agency shop" in the US), requires the employer to collect union dues from all employees irrespective of membership in the union. This arrangement effectively forces all employees to "support" the union and clearly provides unions with a substantial amount of security. Once a work site is certified, the union's funding from all employees is protected by law.

Legislation protecting unions varies across jurisdictions. For example, in some provinces employers collect union dues only when employees authorize the deduction, whereas in others, Rand formula provisions must be bargained for. Under federal labour legislation, however, the Rand formula is incorporated into a contract simply at the request of the union. It is a legislated "right" granted to unions without the need for bargaining. The underlying belief behind this legislation is that employees benefit from collective bargaining and therefore those who choose not to join the union should not be allowed to "free-ride" by not paying dues. The legislators apparently are not as concerned with the fact that in the absence of a union with exclusive representation powers, higher skilled individuals may be able to negotiate a superior wage and benefits package, and therefore are actually worse off under a collective bargaining regime.[There is also the issue of individual rights and freedoms-see Roger Bedard, "Closed Shop Provisions Violate Canadian and Provincial Charters of Rights and Freedoms," in Fazil Mihlar, ed., Unions and Right-to-Work Laws: The Global Evidence of Their Impact on Employment, The Fraser Institute (forthcoming); and the Supreme Court decision on Lavigne vs Ontario Public Service] Moreover, the legislators are also not concerned, apparently, about the possibility that union power may actually have a negative effect on employees.

Amendment: prohibition on replacement workers

The most controversial proposed amendment to the Canada Labour Code introduces provisions for banning replacement workers. Although Bill C-66 does not propose a total ban, the present wording is very broad:

No employer or other person acting on behalf of an employer shall use, for the purposes of undermining a trade union's representative capacity the services of a person who was not an employee in the bargaining unit on the date on which notice to bargain collectively was given and was hired or assigned after that date to perform all of part of the duties of an employee in the bargaining unit on strike or locked out. (s. 94 (2.1))

Currently, the Canada Labour Code allows for the use of replacement workers. What is especially puzzling about the proposed amendment is that the Sims Report (emerging from the task force report that preceded the drafting of Bill C-66) explicitly recognized that "[r]eplacement workers can be necessary to sustain economic viability of an enterprise in the face of a harsh economic climate and unacceptable union demands" and went on to say that "[i]f this option is removed, employers will begin to structure themselves to reduce their reliance on their permanent workforces for fear of vulnerability, to the detriment of both workers and employers alike."[Andrew C.L. Sims, Rodrigue Blouin, and Paula Knopf, Seeking a Balance: Canada Labour Code Part 1 Review, p. 130]

In the event of a work stoppage, it may be necessary for the employer to continue operating in order to maintain the viability of the business. It is interesting to note, however, that employers typically are reluctant to do so. In federally regulated industries, external replacement workers were hired in only 12 of the 48 work stoppages between 1991 and 1994.[Seeking a Balance: Canada Labour Code Part 1 Review, p. 128] Employers have a self-regulating incentive to avoid using replacement workers if at all possible because doing so undermines employee relations and morale. Although most of the employers coming under the umbrella of the federal labour code do not use replacement workers, the ability to do so, or to threaten to do so, provides balance in the bargaining process.

While the Sims Report recognizes there may be a need to ban replacement workers in some "exceptional circumstances," it confirms that a prohibition on replacement workers undermines labour-management relations and business planning. Most significantly, it fundamentally skews the balance of negotiating power. This in turn results in companies reducing their reliance on a permanent workforce, and seeking more business-friendly locations. Furthermore, there is substantial evidence indicating that labour legislation skewed in favour of unions reduces investment and employment. Through increased bargaining power, even a partial ban on replacement workers will contribute to greater union wage premiums and decreased employment growth.

Amendment: organizing off-site workers

A second amendment, intended to assist unions in certifying off-site workers, will also reduce the attractiveness of the overall economic climate for potential investors. Under the proposed legislation, an employer would be obliged to provide a union representative with all the names and addresses of "employees whose normal workplace is not on premises owned or controlled by the employer" and authorize communication with these employees for the purposes of soliciting trade union memberships or other union activities (s. 109.1(1)). Subsection (3) of the same section extends this "assistance" in the union's drive for certification even further by providing that subject to conditions set by the Labour Relations Board, the trade union be allowed to "use any electronic communications system that the employer uses to communicate with the employees."

The Canadian Chamber of Commerce has correctly pointed out that many employees in the federal sector currently work "off-site" (letter carriers, truck drivers, couriers, salesmen, pilots, seamen, pipeline workers, etc.) and yet they have access to collective bargaining. The Chamber argues that "nothing in principle distinguishes the new category of off-site worker which the Task Force has in mind-the homeworker or teleworker-to warrant creating an extraordinary new Board access order.["Canadian Chamber of Commerce, "Submission to the Task Force-Review of the Canada Labour Code (Part 1)," p. 5] The recent and rapid increase in the number of homeworkers and teleworkers reflects the desire of both employers and employees to increase flexibility in working relationships. In addition to serious privacy concerns, forcing employers to assist in certification drives may lead to higher rates of unionization in areas where employment flexibility is a highly desired attribute.

Amendment: air transport contractors

Section 47.3 of Bill C-66 proposes to restrict the competitive bidding process. The clause would force succeeding contractors of pre-boarding airport security services to pay its employees at least as much as the previous contractor. If passed, this amendment imposes existing wage rates on what would otherwise be a free collective bargaining process, and thus limits the bidding process for security services.

As drafted, the legislation explicitly applies to a very narrow sector. However, an additional provision allows the Governor in Council (Cabinet) to designate other federally regulated industries as subject to the same restriction of wage rates paid by previous contractors applying to all successive contracts for the same services. In effect, wage floors could be legislated in any industry within federal jurisdiction. Most significantly, this type of legislation potentially allows Cabinet to interfere with freely negotiated contracts and, in so doing, inhibits the collective bargaining process by legislating a minimum pay scale within different industries.

This proposed legislation would distort labour market activity by legislating wage inflexibility in specific sectors of the economy. Practically, the provision would prove exceedingly difficult to administer because it is necessary to determine if the new services provided are in fact the same as the services provided under the original contract. For example, is a service that changes significantly as a result of a technological improvement actually the same as a service that previously was done manually?

Amendment: remedial certification

Under Bill C-66, when the Industrial Relations Board (also established by the Bill) believes an employer has acted unfairly, it may certify a trade union despite lack of evidence of majority support (s. 99.1). This clause is clearly intended to stop employers from using intimidation and coercion to prevent certification of a non-union site. Though protection of an employee's freedom to decide whether or not they wish to join the union is important, this amendment would grant the Board too much discretionary power.

One possible explanation for these union-friendly amendments is the influence unions wield thanks to their funding and coordinated activities. In contrast, those who bear most of the economic costs of unions-the unemployed-have neither resources nor an effective voice. [For a public choice explanation see James M. Buchanan, Robert D. Tollison, and Gordon Tullock eds, Towards a Theory of a Rent Seeking Society, College Station: Texas A&M University Press, 1980] Arguments in support of unions are largely based on the readily identifiable benefits that accrue to union members. The case against trade unions, on the other hand, is based largely upon the fact that these benefits are extracted through monopoly power and produce distortions and rigidities in the labour market, and increase unemployment. The latter argument is more difficult to make because, while the societal costs may be widespread and potentially much greater than the benefits, they are indirect and diffused. In contrast, the benefits are concentrated and readily linked to membership in a union.

The following section makes the more difficult argument against unions and shows why any legislation that furthers union power and density will be at the expense of perpetuating or increasing labour market rigidities and high rates of unemployment.


Unemployment: Theoretical Considerations

In order to fully understand the relationship between unions and unemployment, and why reforms that liberalize labour markets are necessary to address the unemployment problem in Canada, it is useful to distinguish between cyclical and structural unemployment. We begin from the proposition that, at any time, a national economy can be characterized by a "natural rate" of unemployment which is simply the long-run equilibrium rate given the structural conditions that exist within the economy. These structural conditions include generous employment insurance and welfare programs, and labour market regulations such as minimum wage laws and rules regarding union certification. In addition, there is also a separate and distinct cyclical component of unemployment that creates fluctuations around the natural rate. [Paul Krugman, "Past and Prospective Causes of High Unemployment," Federal Reserve Bank of Kansas City Economic Review, Fourth Quarter 1994, pp. 24-43]

The difference between structural and cyclical unemployment is simple. The former is determined by the structure and economic incentives that exist within labour markets, and the latter arises from fluctuations in economic activity. When recessions occur, cyclical unemployment rises because firms lay off workers in response to a decline in demand for goods and services. As noted at the outset, this type of unemployment cannot account for Canada's ongoing high unemployment rate. In prosperous times companies expand, new companies emerge, and employment levels rise accordingly. The distinction between structural and cyclical unemployment is important because policy prescriptions should aim at the origins of the underuse of labour resources.

The natural rate of unemployment

The idea of a natural rate of unemployment was originally articulated in separate papers by professors Edmund Phelps and Milton Friedman.E.S. Phelps, "Money Wage Dynamics and Labor Market Equilibrium," Journal of Political Economy, no. 76, August 1968, pp. 678-711; and Milton Friedman, ["The Role of Monetary Policy," American Economic Review, no. 58(1), March 1968, pp. 1-17] The term "natural" does not imply that it is in some way good, or that it is the efficient rate where unemployment results only from people changing jobs. Rather, "natural" is the rate to which the economy will move, given the structural characteristics that exist in the commodity and labour markets. Determinants of the natural rate include "market imperfections, stochastic variability in demands and supplies, the costs of gathering information about job vacancies and labour availability, the costs of mobility, and so on[."Friedman, "The Role of Monetary Policy," p. 8] While it is impossible to capture all distortions and market imperfections, empirical work has focused on the mismatching of skills, demographics, unemployment insurance (and other social welfare schemes), minimum wages, payroll taxes, and union density as the main determinants of the structural or natural rate of unemployment. [See OECD Economic Surveys Canada 1996, pp. 157-160; and The OECD Jobs Study: Evidence and Explanations Part II for a discussion of the structural determinants of unemployment; and Herbert Grubel and Michael Walker eds., Unemployment Insurance; and Michael Walker, "The Obvious Solution For Unemployment Insurance," Fraser Forum, December 1994, p. 11-13] The essence of the natural rate is that once all labour market distortions have been considered, the laws of supply and demand determine the equilibrium rate of unemployment.

Like any other commodity, the quantity of labour demanded and supplied depends upon price-increase the wage rate and the demand for labour falls. According to the textbook theory of labour, the demand curve is downward sloping because of the law of diminishing returns. In the short-term, businesses operate with a fixed amount of machinery, computers, and work space. Initially, the first workers in combination with capital equipment are highly productive and output increases with the addition of each employee. At some point, however, output per worker begins to decline as the ratio of capital to labour decreases. Each additional worker still produces a positive, but diminishing, amount of output.

Firms, which are in business to earn profits, implicitly compare the value of output produced by an employee with the cost of hiring that employee. If the wages a firm must pay are lower than the value of an additional employee's output, the firm will make additional profits by hiring additional labour. On the other hand, if wage costs exceed the value of an additional employee's product, the firm will not hire that worker because doing so will reduce profits. Profit maximizing firms will hire additional workers to the point where the value of their production is just equal to the wage the firm must pay that worker.

In a competitive environment, the wage a firm must pay to attract employees is determined in the labour market. If the going wage rate is $10.00 per hour, a firm will hire employees to the point where the value of additional production from a worker approaches $10.00. If other factors, such as productivity and product prices remain the same, and, for some reason, the wage rate rises to $12.00, the firm will respond by reducing its workforce, because now it must pay some employees more than the value of their marginal product. The short-run response of the firm is to reduce its workforce until the output of the last employee is at least equal to the value of product he or she produces. Because firms throughout the economy will also be affected by the wage increase, the overall unemployment rate will increase whenever wage rates rise.

Even if the increase in wages above equilibrium rates is short term, it can have a lasting impact on the unemployment rate. In response to the sudden wage increase, firms may substitute machinery and equipment for labour. With time to adjust, firms respond to the fact that labour has become relatively more expensive than capital by investing in more equipment. [Herb Grubel and Joseph Bonnici, Why Canada's Unemployment Rate is so High, The Fraser Institute, August 1986; and Ron Parker, "Aspects of Economic Restructuring in Canada, 1989-1994", Bank of Canada Review, Bank of Canada, 1995]Once the investment has been made, the firm's demand for labour has been reduced until the capital equipment wears out.

In terms of labour market operations, the basic theory sketched above is very robust. There is a clear relationship between employment rates and wage rates. As a recent comprehensive study by the OECD states, "[t]here is substantial evidence that the demand for labour is negatively related to labour costs in the longer run: lower wages induce higher employment at a given level of output." [OECD, The OECD Jobs Study: Evidence and Explanations Part II, p. 1] The study goes on to conclude both "theory and empirical evidence suggest that lower labour costs stimulate employment in the private sector." [The OECD Jobs Study Part II, p. 51] The relevant question then is how wages get, and remain, too high. An excess supply of labour should lead unemployed people to compete for job vacancies that do arise, and thus push wages down.

This competitive process may be impeded, however, by a number of factors. For example, payroll taxes increase the cost of labour (even if temporarily); there are significant search costs associated with looking for work; firms may pay wage premiums voluntarily in order to retain workers and reduce training costs; and unions, with monopoly bargaining power, are frequently able to negotiate wage settlements in excess of market rates. In addition to the direct effect on wages, unions create other rigidities and distortions in the labour market. Thus the natural rate of employment equilibrium emerges from a competitive labour market operating with these distortions and imperfections.

Unions and the labour market

For members, one of the primary benefits of a union is an increase in bargaining power with employers. With a collective voice and, more importantly, the threat of a strike, unions aim to negotiate higher wages for members than they otherwise would be able to secure as individuals. Unions also have been able to bargain for greater job and income security for their members than non-unionized individuals have been able to attain. In the past, the collective voice of a union has also improved working conditions substantially. All of these are laudable goals and valuable achievements. Overall, it is generally accepted that union membership benefits the majority of its members. [Although as noted previously, collective bargaining may result in higher skilled employees receiving lower wages than they would otherwise be able to negotiate]

However, it is also important to consider the costs. Negotiating higher wages is of course desirable for incumbent union members, but it necessarily has implications for the rest of the economy. When trade unions are successful in raising wages above market equilibrium, affected firms respond i) by raising their price and selling fewer products or services (indirectly reducing employment) or ii) if they cannot raise prices because of competition, they will reduce the amount of labour employed. There is also a third effect that may arise if prices cannot be increased fully to offset increased labour costs and/or the workforce cannot be reduced accordingly. Because unions are able to capture some of the payment to capital in the form of higher wages, firms may choose not to invest additional capital in a plant or industry as higher returns are available elsewhere. Over time, employment will be relatively lower.

Union wage premiums

In order to attract and keep employees, a firm must pay the going market wage. If a work site is certified by a union, then wages are determined through collective bargaining. To ensure membership remains attractive, unions have an incentive to bargain for a mark-up above the market wage. The extent to which wage settlements depart from market wages depends on the relative strength of the union, the heterogeneity of the workforce, and the type of industry the firm is in. In the bargaining process, unions use strikes, or simply the threat of strikes, to negotiate higher wages for all union members. At the same time, firms enter contracts to pay more than market wages in order to avoid the high costs of having the whole operation shut down. The capacity for unions to raise wages is increased under federal labour laws as membership is effectively mandatory, and monopoly bargaining rights entrenched under Rand Formula provisions. As with prices in the product market, the creation of a monopoly in the labour market results in higher wages.

Empirical measurement of wage differentials between union and non-union firms confirms that unions are able to negotiate premiums. Table 2 shows the average wage rates for union and non-union workers in Canada.

Clearly, the average wages for union members are significantly higher. In both 1984 and 1990 the average union member received 30 percent more in remuneration than his or her non-unionized counterpart. It should be noted, however, that straight averages may not accurately reflect wage differentials. Unionized occupations may be relatively high-skilled and members may have more education than workers in the non-unionized sectors. The higher wages in the unionized sector may, in part, reflect additional training and differences in expertise. For this reason, many studies adjust for differences in human capital in order to more accurately estimate wage differentials.

Studies that compensate for differences in human capital still find that unionized workers receive significantly higher wages than non-unionized workers, although the magnitude of the adjusted premium is somewhat smaller than unadjusted measures. In the United States, for example, after allowing for skill differences, compensation for unionized workers was found to be 18 to 20 percent higher than for non-unionized workers. The same study reports a similar wage premium for unionized employees in Canada, while the differential in the United Kingdom, at 10 percent, is somewhat lower. [David BlanchFlower and Richard Freeman, "Unionism in the United States and Other Advanced OECD Countries," Industrial Relations, vol. 31, no. 1, 1992] The consensus estimate reported by the OECD for the union/non-union differential in the United States was also between 10 and 25 percent.[The OECD Jobs Study Part II, p. 11]

All other things being equal, the wage-raising power of unions lowers employment in unionized sectors or industries. It should be noted, however, that the negative employment effect in the unionized sectors might be mitigated if unionized companies were more efficient than comparable non-unionized enterprises.[The potential for productivity differences to compensate for high wages is frequently made in the literature and is articulated in The OECD Jobs Study Part II, p. 12]f unionized workers actually produced more per hour than non-unionized workers, then the value of output per worker in the unionized sector would be greater and, as labour theory suggests, firms would be willing to pay higher wages at the same level of employment. Studies on productivity, however, do not reveal any differences. In fact, in Canada (as discussed below), there is evidence that unionized firms are actually less productive. Within a broader multi-country perspective, the OECD study concludes: "With no clear evidence of a productivity offset, the wage-raising power of unions is thus likely to depress employment in the unionized sector or in sectors where the bulk of employees are covered by multi-employer collective contracts." [The OECD Jobs Study Part II, p. 13]

In addition to reducing employment in unionized sectors, there is evidence that unions also affect employment in non-unionized sectors. If wages were fully flexible in the non-unionized sectors the overall unemployment rate may not necessarily increase in the presence of high rates of unionization. Workers that were displaced from union enterprises as a result of high wages could theoretically become gainfully employed within the non-unionized sector if this part of the labour market was free to adjust. In reality, however, governments impose minimum wage laws which restrict the downward adjustment of wages while other social policies, such as unemployment insurance, distort labour market operations and act as wage floors. Moreover, the mere existence of unions and the concomitant threat of unionization can affect wages paid in the non-unionized sector.

Wage impacts in non-union firms

Many non-unionized firms pay above market wages in order to avoid having workers unionize. The "threat factor" results in higher wages in the non-union sector because for non-unionized firms it is less costly to pay a wage premium and minimize the risk of union certification than it is to incur the high costs associated with unionization. A number of studies indicate that a higher proportion of unionization in an industry generally produces larger non-union wage premiums. [Barry Hirsch, Unionization & Economic Performance: Evidence on Productivity, Profits, Investment and Growth, Public Policy Source No. 3, The Fraser Institute, June 1997; Grubel and Bonnici, Why Canada's Unemployment Rate is so High; and Donald J. Daly and Donald C. MacCharles, On Real Wage Unemployment, Focus No. 18, Vancouver: The Fraser Institute, 1986]

That the degree of wage matching within industries depends upon the size of the firm. Large, non-union employers tend to match union wage rates irrespective of union density in their industry; mid-sized firms match union pay scales when the threat of unionization is significant; and small non-union employers pay wages that are below union rates at all levels of union concentration. [Michael Podgursky, "Unions, Establishment Size, and Intra-industry Threat Effects," Industrial and Labour Relations Review, vol. 39, no. 2, Jan. 1986] Generally, large firms are more attractive for union organizing drives, and so it should not be surprising that the large firms are the ones that match wages. Threats result in higher wages in non-union sectors, and therefore strong unions or high rates of unionization lower employment in a broader segment of the economy. [Sometimes governments introduce regulations that affect wages in the non-union sector. For example B.C.'s Fair Wage Policy requires all highway contractors to pay union wages even if they are non-union firms]Enacting legislation that makes it easier to certify workplaces will only magnify companies' concerns about drives to unionize and reduce employment growth as firms react to the potential threat of unionization.

Unions, profits and productivity

The effect of unions on profits is the largest concern of management and investors. The fact that firms so strongly resist unionization suggests that management believes that union certification reduces profits. Given that unionized employees are paid significantly more than non-unionized ones, this intuition seems justified-higher labour costs result in higher production costs. As well, the restrictive work practices and additional fringe benefits that unions frequently negotiate also increase costs. Empirical evidence supports both the theory and intuition. Studies find that unionized firms are generally less profitable than non-unionized firms. An early example is a study that used microeconomic data on 902 businesses over the period 1970-80 and examined the rates of return on investment. After controlling for differences in industry structure and industry labour markets, the author found that returns on investment are 19 percent lower in unionized firms relative to the mean return of non-unionized firms. When measured by the return on sales, unionized firms are 18 percent less profitable than their non-union counterparts. [Kim Clark, "Unionization and Firm Performance: The Impact on Profits, Growth, and Productivity," The American Economic Review vol. 74, no. 5, December 1984, pp. 893-919]

Another study takes a different approach and uses the efficiency of equity markets to empirically test the hypothesis that unions adversely affect profits. If companies with unionized work forces are less profitable, then the stock price of the company that has just been unionized should reflect this difference relative to other companies. After examining 253 union certification elections for firms listed on the New York Stock Exchange, the authors found that on average, unionization is associated with a reduction in profitability. Specifically, certification of firms resulted in a 4 percent reduction in equity value. Interestingly, in firms where workers petitioned for an official certification election but subsequently voted not to unionize, equity values still declined-but only by approximately 1 percent. [Richard Ruback and Martin Zimmerman, "Unionization and Profitability: Evidence from the Capital Market," Journal of Political Economy vol. 92, no. 6, December 1984, p. 1,137]

A more recent study of the Canadian manufacturing sector also finds that unions have a negative effect on profitability. After accounting for industry structure, the study found that adverse impacts of unions on profits depend upon the degree of competitiveness of the industry in which a firm operates. In highly competitive manufacturing industries there appears to be no negative effect on profitability. In less competitive industries, however, unions are able to extract an increasing proportion of the firm's profits. The less competitive the industry, the larger the proportion of incremental profits unions are able to extract. [Pasquale Laporta and Alexander W. Jenkins, "Unionization and Profitability in the Canadian Manufacturing Sector," Industrial Relations vol. 51, no. 4, 1996; and Barry Hirsch, "Union Coverage and Profitability Among U.S. Firms," Review of Economics and Statistics 73, February 1991, pp. 66-77]

Unions may also have a negative impact on firm productivity. The empirical evidence on this measure, however, is somewhat mixed. Initially, studies that examined productivity performance using production functions found that unions sometimes provided small positive benefits and sometimes small negative effects. [Charles Brown and James Medoff, "Trade Unions and the Productive Process," Journal of Political Economy vol. 86, (June) 1978, pp. 355-78; and Richard Freeman and James Medoff, What Do Unions Do?] The explanation for positive union effects on productivity was typically that unions reduce turnover, provide efficient structures in workplaces and are coupled with appropriate institutional responses from management. To some extent this may be true.

More recent research, however, has contradicted this. Barry Hirsch provides a comprehensive review of the productivity literature which reconciles these apparent inconsistencies. [Hirsch, Unionization & Economic Performance.]The differences in findings are largely attributable to different estimating techniques and different industry structures. One important factor unaccounted for by studies reporting positive productivity impacts is that higher union wages reduce profits and knock some firms out of business. These should be included in estimates of union impacts on productivity, but typically are not, as they no longer exist. Ultimately, Hirsch concludes there is no evidence for a positive union effect on productivity and that in the absence of such productivity increases higher union compensation implies lower profits and lower investment.

Unions and investment

Labour relations are a particularly important consideration for businesses determining where to build or move their operations. Because of the economic effects of unions, firms have an incentive to avoid jurisdictions with high rates of unionization in their industry. Other factors remaining the same, companies will invest in regions with lower union densities and business-friendly environments. In a recent comprehensive study, Thomas Holmes empirically tested the hypothesis that labour regulations and the general business climate influence where US manufacturing firms invest. [Thomas Holmes, "The Effects of State Policies on the Location of Industry: Evidence From State Borders," Federal Reserve Bank of Minneapolis Research Department, Staff Report 205] By examining the concentration of manufacturing entities in counties along borders where states with right-to-work laws are contiguous to states without right to work laws, Holmes was able to test the hypothesis that labour regulations affect the location of investment.

The results are striking. They show a large and statistically significant discontinuity in manufacturing activity along policy-change borders-manufacturing employment increasing by fully one-third when one crosses over the border from the "anti-business" side to the "pro-business" side. [Holmes, "The Effects of State Policies on the Location of Industry," p. 3] By testing adjacent locations, the effects of many other potential influences such as weather and varied access to transportation networks are eliminated. After testing for the influence of other factors, Holmes also concluded that similar discontinuities do not occur within regions with the same labour legislation. As well, he conducted statistical tests to determine if the state labour policies also affect industry locations further from the border. He found that the differences detected at the border do not diminish as business location sites move away from the state lines delineating policy changes.

These findings are particularly interesting in light of the fact that there has been a dramatic shift of manufacturing activity from states that do not have right-to-work laws to states with right-to-work laws. Between 1947 and 1992, employment in manufacturing increased by 148 percent in jurisdictions that protect an individual's right not to be forced to join a union as a condition of employment. By comparison, in states without such laws, there was virtually no growth in manufacturing employment. [Holmes, "The Effects of State Policies on the Location of Industry," p. 2]Although he notes that the differences in labour legislation may not necessarily have caused the shift in manufacturing activity away from the manufacturing belt, Holmes does conclude there is a clear relationship between the general business climate and the divergent growth patterns, and that variation in state policy is an important factor in accounting for the redistribution of manufacturing jobs. Considering that most of the states with right-to-work laws had passed the legislation by the late 1940s and early 1950s, it is reasonable to infer that the labour legislation has had a significant impact on where firms have chosen to invest in the post war era.

Impact of unions on employment

In the context of this study, the most important concern is the impact unions have on employment. Ultimately, all of the wage, profitability, productivity, and investment effects of unions discussed above should manifest themselves in lower employment growth. Is there any empirical evidence of a negative employment effect within unionized sectors as well as the economy in general? Are unions a contributing factor to the high rate of unemployment in Canada?

A 1993 study by Professor Richard Long examined employment growth in Canadian manufacturing and non-manufacturing firms and found strong evidence that unions do depress job growth. [Richard Long, "The Effect of Unionization on Employment Growth of Canadian Companies," Industrial and Labour Relations review vol. 46, no. 4, July 1993, pp. 691-703] He traced the changes in the workforces of 510 firms between 1980 and 1985, 270 of which were unionized. Summary statistics of the number employed in the sample of manufacturing and non-manufacturing firms are presented in table 3. Over the five year period, total employment in the manufacturing sector declined by approximately 6 percent. On a union-non-union basis, however, growth patterns were very different. Overall, employment in the unionized firms fell by nearly 13 percent, which is in sharp contrast to the 23 percent increase in employment in the non-union manufacturing sector. A similar pattern was also found in the non-manufacturing sector. As the lower part of the table shows, employment in the unionized firms remained essentially the same, growing by less than 1 percent, while non-union firms in industries other than manufacturing recorded increases in the total number of employees of more than 15 percent.

The last columns in the table represent the mean and median growth rates of employment in individual firms. In the manufacturing sector, union firms grew on average by 10.5 percent. Their non-union counterparts grew significantly more, at an average rate of 80.7 percent. The reason for the differences between the change in total employment and average growth rates is that large unionized firms experienced employment losses, while smaller union firms experienced employment gains. Similarly, the fact that on average non-union firms nearly doubled their work forces while total employment only increased by 23 percent means the smaller firms grew more than did the large non-union firms. In the non-manufacturing sector, the difference between the employment performance of union and non-union firms was almost the same as the difference in the manufacturing sector with growth rates of 10.3 percent and 78.6 percent respectively. The same discrepancies between employment growth indicators (the increase in total employment and average growth rates) also exist in the non-manufacturing sectors because of the more rapid expansion of small firms. Clearly, there is evidence that employment growth is lower in the unionized sector of the economy. Does this employment effect transcend to the non-union sector and reduce overall employment throughout the economy?

One indirect test of the effect of unions on total employment is to isolate the determinants of the the long-term unemployment rate over the last 20 years. One such study, conducted by economists at the Bank of Canada, employs this general methodology. [Denise Côté and Doug Hostland, "An Econometric Examination of the Trend Unemployment rate in Canada," Bank of Canada Working Paper 96-7] Using sophisticated statistical techniques, the authors test a number of different factors including demographics, minimum wages, payroll taxes, and unemployment insurance programs; and estimate the unemployment rate trend over time. The study concluded that long-run fluctuations in the Canadian unemployment rate between 1955 and 1994 "can best be explained by two structural factors: the degree of unionization in the labour force and payroll taxes." [Côté and Hostland, "An Econometric Examination," p. 35] The results suggest that union density has been an important determinant of the unemployment rate in Canada and that the general increase in unemployment is explained, at least in part, by unionization.

Canada's unemployment problem: structural in nature

For decades, successive economic recoveries have not seen the unemployment rate return to its pre-recession level. Accordingly, there is agreement among economists that the natural unemployment rate in Canada has risen significantly. There is less agreement, however, as to the magnitude of the change and what the current natural unemployment rate actually is. Although it may be difficult to find a consensus estimate for the current natural rate, the behaviour of the actual unemployment rate (the cyclical component and the structural component) provides insight.

Most significantly, the measured unemployment rate has not been below 9 percent for 6 years and has remained higher than 7.5 percent for more than 16 years. Given this basic observation, we can rule out Finance Minister Paul Martin's suggestion that the natural rate might be as low as 6.5 percent and that 5 percent might be achievable, without substantial labour market reforms. [The Globe and Mail, Monday, February 24, 1997, B1] Realistically, the natural rate is more likely in the 8.0 to 9.4 percent range. Average estimates from several studies conducted between 1992 and 1996, each of which used sophisticated modelling techniques, produced an unemployment rate estimate of 7.9 percent-at the lower end of this range. A rule of thumb measure, [The OECD uses an 8 year moving average to proxy shifts in the natural rate of unemployment. For details, see OECD Economic Surveys: United Kingdom, 1995, chapters 1 and 4]and the fact that the actual rate remains close to double digits, suggests that an estimate near or above the high end of the range is more appropriate.

Although rising unemployment rates over the last three decades has been common in most industrialized countries, Canada's unemployment record ranks among the worst of the OECD countries. In fact, much concern and analysis has been produced by the persistent gap that has emerged between the unemployment rates in Canada and the United States. Just two decades ago, both countries had the same proportion of people without jobs. Beginning in the 1970s, and continuing throughout the 1980s, however, the growth in Canadian unemployment outstripped that in the US. Following the 1991 recession, the differential widened to a high of 4 percent, where it has remained to date.

A comparison with unemployment rates in other OECD countries reveals a similar pattern. Figure 1 depicts Canada's relative unemployment performance to total unemployment in 18 OECD countries and the US, using standardized rates.[The standardized unemployment rates are produced by the OECD and give the number of unemployed persons as a percentage of the civilian labour force. The definition of unemployment conforms with the definition adopted by the 13th Conference of Labour Statisticians. See the OECD's Quarterly Labour Force Statistics for a description of sources and methods. The total unemployment rate is the proportion of the labour force in 18 OECD countries]What is striking is that throughout the period, unemployment in Canada has remained well above the combined OECD average unemployment measure which includes most western European countries. In 1981 the gap declined to one percent, although it subsequently grew to as much as 4 percent, and today there remains a two percent differential between the unemployment rate in Canada and the OECD benchmark.

With the post 1991 economic recovery came a marginal improvement in employment prospects for jobless Canadians. After approximately two years of decline, however, the unemployment rate levelled off at approximately 9.6 percent. In fact, in the last two quarters of 1996, Canada's unemployment actually rose slightly to 9.7 and 9.9 percent respectively. By way of comparison, other OECD countries have seen significant reductions in the number of jobless over the same period. New Zealand, for example, registered a 4 percentage point reduction in its unemployment rate, from 10.3 percent in 1991 to just under 6 percent by the end of 1996. Over the same period, the British unemployment rate fell significantly and is 2.2 percentage points lower than the equivalent Canadian measure.

It is worth noting that seemingly small reductions in the Canadian unemployment rate of one or two percent represent a substantial improvement in total economic well-being-a single percentage point decline in the unemployment rate means another 150,000 Canadians would become economically productive.

Though the official Canadian unemployment figures are high, there are two reasons to believe the real rate is even higher. First, there are a large number of discouraged workers not captured by the unemployment statistics because they are not actively seeking work. Anyone who has given up hope of securing a job is not considered part of the labour force and thus, statistically speaking, is not "unemployed," even though they would probably accept an offer of employment. On average, in 1996, the number of discouraged workers in Canada was approximately 306,000. [Statistics Canada, The Labour Force, cat. 71-001, Annual Averages 1996. The survey identifies persons who looked for work in the past six months but not in the four weeks preceding the survey. This estimate is conservative because it does not count long-term discouraged workers (those who did not look for work in the previous six months)]

The second consideration, and one more directly linked to rigid labour markets, is the increasing prevalence of involuntary part-time employment. In an attempt to avoid incurring the additional costs associated with full time employment, firms are increasingly opting for more flexible part-time arrangements. For workers, however, part-time employment masks some of the unemployment problem because anyone who is involuntarily working part-time is counted as "employed." [According to the OECD (OECD Economic Surveys: Canada 1996, p. 62), Canada ranks third in the occurrence of underemployed males and fourth for females who would prefer to be employed in a full time job]While many Canadians prefer to work part-time because of family responsibilities, recreational activities, etc., the 800,000 Canadians who would prefer to work full-time represents excess labour the economy is unable to use. [Statistics Canada, Labour Force Annual Averages 1995, cat. 71-220]

Frequently, the federal government claims that while job creation in recent years has been reasonably good, other factors such as changing demographics and a rising participation rate are keeping the unemployment rate high.[According to their Agenda: Jobs and Growth policy document, "Canada has achieved an excellent rate of job creation ... but our labour force has grown even faster, causing average unemployment to rise." The same document goes on to state that "Canada's unusually high rate of labour force growth has been due to: (i) demographic factors, notably [the] baby boom generation and, more recently, high rates of immigration, and (ii) the strong growth of female participation." Agenda: Jobs and Growth, A New Framework for Economic Policy, p. 18] It is true that Canada has had a strong record of employment growth over the last 20 or 30 years, but in recent years the rate has been unusually low. The growth in the labour force has followed a similar downward pattern.

If a growing labour force is one of the main reasons unemployment rates have remained high, we would expect the growth rates in the 1990s to reflect this rapid expansion. Instead, as Figure 3 reveals, the opposite is true. In the 1990s, the workforce grew by slightly more than one percent, whereas during the 1970s it increased 3.0 to 3.5 percent annually, and by 2 percent during the 1980s.

The slow growth of the labour force may reflect, in part, the poor employment prospects over the same period. As figure 4 shows, job expansion in the 1990s was very slow to pick up, has averaged only 1.5 percent annually, and already appears to be declining. By comparison, the growth in total employment in the two previous recoveries was sustained at annual rates of nearly 3 percent per year for 5 or 6 years. Job growth for the narrower measure of full-time employment (figure 5) also displays a similar pattern to that of total employment, except that full-time employment grew more slowly and has fallen off more dramatically.

The rising and sustained high rates of unemployment in Canada suggest that much of the problem is structural in nature. Poor employment growth in recent years is a further testament to the fact that there are worsening structural impediments to job creation. The implication of a high structural rate of unemployment is that governments must focus their policy measures on the institutions and rigidities in the labour market. Governments cannot have any lasting influence on the unemployment rate through expansionary monetary or fiscal policy because it is the institutions and imperfections within the labour market that determine the long-run unemployment rate.

What impact might such liberalization of labour markets have? While difficult to determine exactly what effect reforms would have because there are so many factors and variables involved, it is clear that comprehensive labour market reform would help reduce unemployment rates in Canada. The United States, which has one of the most flexible labour markets in the world, has an unemployment rate below 5 percent. Interestingly, the degree of unionization emerges as one of the most prominent differences between the two labour markets. In Canada, 37.5 percent of the workforce is unionized compared to only 15.5 percent in the U.S. One study estimates that the difference in unionization accounts for roughly 2.5 percent of the spread in unemployment rates between the two countries. [Results of a study conducted by Lawrence Summers, Deputy Secretary of the U.S. Treasury, reported in David Henderson, "Canada's High Unemployment Rate is No Mystery," The Wall Street Journal, February 7, 1997]

The experiences of Britain and New Zealand provide perhaps the best evidence that unemployment would be reduced by reforms limiting union security and protecting an individual's right to join or not join a union. In the UK, major industrial relations reform has led to a more responsive labour market and an unemployment rate of 7.0 percent [OECD, Main Economic Indicators, July 1997] (which 10 years ago was considered as unattainable in Britain as is a 6 percent rate in Canada today). In New Zealand, similar reforms produced a 4 percentage point decline in unemployment.


Case Study: Britain

Until recently, the pattern of the unemployment rate in Britain resembled that in Canada. In 1970 only 3 percent of British workers were unemployed. By the mid 1970s this proportion had increased to 6 percent, and by 1984 fully 13.5 percent of the labour force was unable to find a job. Like Canada, the unemployment rate declined slowly following the recession of the early 1980s and then increased again towards the end of the decade. In more recent years, however, there has been a divergence of the two unemployment rates. In the recession of the early 1990s, the unemployment rate in the UK remained below that in Canada and has recovered much more rapidly since. Comparable figures indicate that the unemployment rate in Britain has fallen steadily to 7.0 percent, while the rate of unemployment in Canada remained above 9.5 percent for three years.

The stronger employment growth in Britain is attributable to the labour market reforms that characterized the Thatcher years (1979-91) in the United Kingdom. The history of labour legislation and its reforms are discussed here because much of the British unemployment problem was attributable to rigidity in labour markets stemming from excess union powers and immunity. It was Milton Friedman's original articulation of the natural rate hypothesis that focused attention on these structural concerns and gave rise to the Thatcher reforms.

The "British Disease"

The "British Disease" reflected the poor labour market conditions that existed in the UK. Symptoms of the ailment were a high number of strikes, low labour productivity, and poor quality products. As union militancy increased, the disease spread. The 1974 coal miners' strike affected the rest of the economy as other industries were forced to scale down to a three day work week because of the reduction in the supply of electricity. The auto industry, which had once been a world leader, was in especially poor condition. The 350 disputes that British Leyland and Ford experienced in a short six month period in 1977 reveal the gravity of the labour problems. [Charles Hanson, "The Economic Impact of Labour Reform: The UK Experience," Paper presented at The Fraser Institute's Right-to-Work Conference, Toronto, June 21, 1996]

Exploiting extended legal immunities which had been re-introduced by the Labour government (1974-79), many public sector unions went on protracted strikes. Public sector strife culminated in 1978-79, during the so-called "Winter of Discontent." A series of lengthy strikes at national railways, the London underground, town halls, trucking operations, docks, schools, hospitals, ambulances, garbage collection facilities, cemeteries, and municipal airports essentially crippled the British economy. [Charles Hanson, The Taming of the Trade Unions, London: Macmillan Publishing, 1991] By the time it was over, the British people were fed up with union militancy.

The 1979 election of Margaret Thatcher's Conservative government reflected this sentiment. The unremitting labour strife was a highly visible manifestation of excessive union power-power that for years had distorted labour markets, reduced investment in the country, and made employers reluctant to expand their payrolls. The visibility and widespread economic impact of the strikes gave Margaret Thatcher the public support she needed to initiate major labour market reforms. In the short term, it was necessary to eliminate the crippling strikes. However, it was clear that to regain its economic independence, more fundamental changes were required for Britain. Labour market rigidities would have to be substantially reduced.

Britain had a long history of national wage setting arrangements. While 53 percent of employees were union members in 1980, it is estimated that up to 83 percent of employees were covered by collective agreements or Wages Councils. [John Addison and Stanley Siebert, "Union Security in Britain," table 1, in Fazil Mihlar, ed., Unions and Right to Work Laws: The Global Evidence of Their Impact on Employment, The Fraser Institute (forthcoming)] Non-union employees covered by collective agreements typically looked to trade union negotiations, as opposed to individual performance, when securing pay increases. The result was that wages across the entire economy were inflexible and reflected neither differing regional economic conditions nor changes within different industries. It also was exceptionally difficult for companies to hire and fire workers. Perhaps most importantly, employee flexibility was almost non-existent, particularly within the British manufacturing sector. Changes in work practices could not be introduced without agreement from a shop steward, which usually was not forthcoming.

Thatcher's step-by-step approach to eliminating excessive union legal privileges resulted in six significant employment or trade union acts being passed between 1979 and 1993. [Addison and Siebert, "Union Security in Britain", table 2] Contrary to many pessimistic predictions, the reform campaign has been an overwhelming success.

History of labour legislation

The origins of the extraordinary power and influence of British unions dates back to the early years of the century. In 1906, the then Liberal government passed the Trade Disputes Act, giving unions and their officials unconditional legal immunities. The Act stated explicitly that:

An action against a trade union, whether of workmen or masters, or against any members or officials thereof on behalf of themselves and all other members of the trade union in respect of any tortious act alleged to have been committed by of on behalf of the trade union, shall not be entertained in any court (s.4(I)).

For three-quarters of a century trade unions were immune from litigation pertaining to civil wrongs, libel, negligence, or strike activity. The widely quoted sentiments of A.V. Dicey succinctly summarize the effect of the legislation: "[it] makes the trade unions a privileged body exempt from the ordinary law of the land. No such privileged body has ever before been created by an English Parliament." [A.V. Dicey, Law and the Public Opinion in England (1914), Macmillan, 2nd edition 1963, p. 468] In light of this near total immunity, it is not surprising that unions grew in strength and militancy.

By the mid 1960s, even the Labour government recognized that unions were too powerful. The most obvious manifestations of their excessive power were unyielding local and unofficial strikes. The 1906 Trade Disputes Act protected the right to strike by allowing workers to breach their employment contracts when engaged in trade disputes, and protected union funds from liability for damages. These powers meant unions essentially could dictate work arrangements and wage increases. When the Conservative government came to power in 1970, it was determined to loosen the stranglehold which unions had on the country, albeit without much success. In 1971, the government introduced the Industrial Relations Act that included provisions for a labour court; labour practices, including unfair dismissal for the first time in British history; a mechanism for recognizing unions which could then form "agency shops"; and legally enforceable collective agreements. [Addison and Siebert, "Union Security in Britain," p. 5]

In practice, however, the new legislation had little effect on labour relations. Few agency shops were formed, and strikes continued to plague the economy. In early 1974, the Conservative government was defeated by the Labour Party who replaced the Industrial Labour Relations Act with the Trade Union and Labour Relations Act (TULRA). This new legislation restored the legal immunities of the 1906 Trade Disputes Act while also retaining the concept of unfair dismissal that had been introduced as part of the Conservative government's reforms.[Addison and Siebert, "Union Security in Britain," p. 5] The combination of the restored immunities and unfair practices were used to strengthen and expand closed-shop unions and contributed to the labour disputes in the Winter of Discontent.

Thatcher's labour market reforms

Today, industrial relations are very different in Britain. Introduced gradually over a 13 year period, the reforms removed many union privileges and protected individual rights, including the right to bargain collectively. The essential changes contained in the successive Acts weakened the power of unions by eliminating their immunities, improved trade union democracy, and most significantly, provided individuals with the freedom to choose whether or not they wish to join a union without fear of being dismissed.

In Britain, closed shops were common and there was no legal protection for a worker's individual freedom. In the period 1974-1979, it was legal to dismiss an employee for not being a member of a trade union. One of the primary objectives of the gradual reform process was to eliminate this violation of workers' fundamental rights. Between 1980 and 1989, dismissal for non-membership was lawful in some instances; by 1990 it was unlawful; and finally, in 1993, the Trade Union and Employment Rights Act went even further by making it illegal to refuse employment on the grounds of non-membership in a trade union. These successive legislative changes have created legal protection for an individual's right to work. New legislation also introduced a number of measures designed to reduce the disruption caused by industrial action and improve the democratic operations of unions. Table 4 highlights the major changes contained in each of the Acts.

The impact of labour reforms

Today, most commentators agree that the British disease has been cured. There is widespread evidence that greater flexibility in industrial relations and labour markets has helped labour markets operate more efficiently since the latter 1980s. This flexibility is partly reflected in the decline in strike activity. Between 1975 and 1979 there were, on average, 2,345 strikes each year. This fell to an average of 334 between 1990 and 1994. In the last two years of the latter period, the number of work stoppages reached record lows of 211 and 205 respectively. The most significant effects of the labour market reforms are at the microeconomic level, where the structural rate of unemployment is determined. Flexibility is evident in the "ease of hiring and firing, greater decentralization of pay fixing and working conditions, wider wage differ- entials according to the skill classifications and greater wage variation across regions." [OECD, OECD Economic Surveys: United Kingdom, 1995, p. 16]

Although union effects on investment in Britain have not been examined as extensively as in the U.S., time series studies have found a strong negative union-investment relationship. Denny and Nickell, for example, found that unionized firms have a 28 percent lower rate of investment than similar non-unionized firms. [Kevin Denny and Steven Nickell, "Unions and Investment in British Industry," Economic Journal 102, July 1992] When reviewing the evidence, Addison and Siebert note that further analysis is necessary, but conclude there is some evidence that the negative effect of unions on investment has become less pronounced following the Thatcher reforms.[Addison and Siebert, "Union Security in Britain," p. 18]

Specific examples highlight this success. In the mid 1980s, Nissan built a major car plant near Newcastle which has yet to experience a work stoppage of any sort.Hanson, ["The Economic Impact of Labour Reform," table 1] Flexible work arrangements are one of the key reasons for the overall success of the factory. The Nissan agreement is comprehensive, but allows for the workers' roles to change and adapt. The opening clause states explicitly: "to ensure the fullest use of facilities and manpower there will be complete flexibility and mobility of employees." [Charles Hanson and Graham Mather, Striking Out Strikes: Changing Employment Relations in the British Labour Market (Institute of Economic Affairs, 1988), p. 23 ] The German car manufacturer BMW also made a sizable investment in 1994 when they bought the British motor car firm, Rover Group. Both of these investments would not have been even considered in 1979. Reflecting upon the decision to purchase the company, the chairman of BMW stated that Britain was "the most attractive country among all European locations for the production of cars [due to] the structural reforms initiated by Margaret Thatcher in the early 1980s." [Hanson, "The Economic Impact of Labour Reform," p. 9]

Profitability and productivity are two fundamental indicators that have been shown to decline under unionism. The negative impact of unions on the profits of British firms is well documented. Even more noteworthy is the marked improvement in corporate profits after 1981 that have been linked directly to the simultaneous decline in the rate of unionization. After reviewing the literature on this topic, Addison and Siebert conclude that the improvement in profitability seems "most clearly allied with the decline in union bargaining power stemming from anti-union legislation." [Hanson, "The Economic Impact of Labour Reform," p. 17]

As is the case in other countries, there is some dispute over the impact unions have on productivity in Britain. In the 1970s unionization clearly impaired the productivity growth of firms. In the 1980s there was some evidence that the union effect on productivity was slightly positive, although probably not enough to compensate for the 10 percent wage premium. There is a need to further assess the impact of unions on productivity in Britain and, in particular, the effect labour reform has had on this output measure. It is interesting to note, however, that one of the most recent studies on productivity finds that productivity growth was the highest in those companies that abandoned the closed-shop and either partially or totally decertified. [Paul Gregg, Stephen Machin, and David Metcalf, "Signals and Cycles: Productivity Growth and Changes in Union Status in British Companies," Economic Journal, 103, July 1993, pp. 894-907]

For Canadian policy makers, the most instructive effect of liberalizing labour markets in Britain has been its effect on unemployment. The Thatcher reforms have resulted both in a decline in the natural rate of unemployment and an increase in the speed of adjustment. As the OECD notes, "a striking feature of the [latest] recovery has been the early drop in unemployment compared with previous cycles." [OECD, OECD Economic Surveys: United Kingdom, 1995, p. 13] Today, estimates of the natural rate of unemployment in Britain range from 5 to 7 percent, a substantial improvement over a rate of more than 10 percent in the mid 1980s.  [The OECD Secretariat adopts a conservative estimate of 7 percent when calculating Britain's output gap. OECD, OECD Economic Surveys: United Kingdom, 1995, p. 23]

The evidence indicates overwhelmingly that the legislative reforms have had a positive effect on the British economy, with the most significant result being a marked improvement in labour market flexibility. The labour market is more responsive to economic conditions because of the decline in the proportion of centralized and collective bargaining which paralleled the decline in union membership. The UK labour market is "now one of the least regulated among OECD countries, as regards terms and conditions of employment, working times, and hiring and firing rules."  [OECD, OECD Economic Surveys: United Kingdom, 1996, p. 88] At the same time, workers are protected from discrimination, unfair dismissal, and a comprehensive framework of health and safety legislation is in place. The most significant and tangible benefit of the reform process is the increase in employment creation and the accompanying decline in the proportion of jobless.


Case Study: New Zealand

Like Britain, New Zealand has undergone a major restructuring of its labour market. New Zealand differs from Britain somewhat, however, because these reforms were introduced as part of a broader process of general economic reform designed to make New Zealand a better place to do business. Prior to 1982, New Zealand had a complex array of subsidies, levies, and regulations, making it one of the most heavily regulated economies in the OECD. The result was a stagnant economy with trade and growth rates among the lowest of all OECD countries.

Faced with near economic crisis due to a massive budget deficit and public debt, the newly-elected reform-minded Labour government initiated a program of rapid market restructuring and reform in 1984. By the early 1990s, farming subsidies were eliminated, quotas abolished, financial markets liberalized, transportation deregulated, and numerous state-run activities privatized. In addition, import subsidies were removed and a broader goods and services tax was introduced subsequently allowing the top rate of income tax to be reduced to 33 percent. With the reforms came renewed prosperity and economic growth.

One of the key aspects of the economic overhaul was reform of the labour market. New Zealand had long been known for its centralized bargaining (usually occupation-based) and inordinate protection and promotion of unions. As was the case in Britain, unions in New Zealand had enjoyed a long history of legally sanctioned privileges. In fact, many of the legislated practices were well in excess of any privileges granted to unions in Britain.

The beginning of the modern labour regulations date back to 1935 when the New Zealand Labour Party made unionism compulsory. For more than half a century, unions were provided a secure flow of membership and union revenue because it was mandatory for all people working in a defined craft area to become members, and no crafts remained non-union. The laws were such that if a craft or occupation was not covered by a union, a few individuals could form a union, register, and henceforth have monopoly rights over all workers in that occupation. [Wolfgang Kasper, Free to Work: The Liberalization of New Zealand's Labour Markets, The Centre for Independent Studies, January 1996, p. 23]

Perhaps even more unusual was the fact that unions could "choose" who they were going to negotiate with. They were allowed to notify a "representative sample" of employers with workers in the newly claimed occupation and negotiate only with those in the sample. [Kasper, Free to Work, p. 23] Under the law, however, all employers in that craft or occupation were required to comply with the agreement in spite of the fact they did not participate in, and often were not aware of, the negotiations. Blanket settlements caused relative rigidities between jobs, and as wages in different industries ratcheted up together, labour resources were misallocated. Artificially high wages also meant that industry wages did not reflect productivity levels. Compulsory arbitration, by which disputes were settled in the government's Arbitration Court without regard to specific regional or business circumstances, also distorted the labour market. [Kasper, Free to Work, p. 24] In short, the legislative framework left labour markets highly inflexible and unresponsive.

In light of the other reforms that were being undertaken, labour market liberalization became more critical. The distortions and wage rigidities that persisted because of the industry-wide bargaining structure were not compatible with the deregulation and liberalization occurring throughout the economy. Firms now had to compete in open markets, yet had no control over their labour costs. Not wanting to erode its support base, the Labour government had been reluctant to include labour markets as part of its economic restructuring. Nevertheless, the pressure for further reforms mounted as it became increasingly apparent that the government's reform package was being sabotaged by a refusal to liberalize industrial relations. In 1990, the National Party was elected on a platform of labour market reform, a key point of distinction from the incumbent Labour Party.

The second wave of reform

The National Party strategy was to bring about rapid labour market reform. Elected in October 1990, it introduced its Employment Contracts Act (ECA) before Christmas, and the Act became law by May the following year. By this time, most other aspects of the economy had been deregulated and were fully subject to competitive forces. The new labour legislation was designed to enhance the flexibility and adaptability of private enterprises so that they could compete effectively in the global market place. The ECA eliminated special union privileges and dramatically altered the rules governing the bargaining process and structures.

The most important effect of the legislation was that it altered fundamentally the employment relationship. Individuals became free to negotiate with employers, in sharp contrast to the mandated industry bargaining that had previously characterized labour relations. The right to associate freely ensured that individuals could make arrangements to negotiate contracts as individuals, as a small unit, as an enterprise group, or even as a multi-enterprise group. Under the amended law, employees and employers may either represent themselves, or choose a union, a lawyer, or labour relations specialist to act as their bargaining agent. The ECA affords no special status to unions, and all agents must have written permission to bargain on behalf of the employee and the employer.

The results of reform

The rationale for the ECA was to make employment relationships operate in a manner similar to those in other sectors of the economy and, in particular, to encourage decentralized bargaining. The objectives of the legislation have been met. Predictions of entrenched high unemployment, "gangster unionism," poor quality jobs, and anarchy all have failed to materialize. Instead, the recent increases in employment and productivity attest to the benefits the Employment Contracts Act has had on the New Zealand economy.

Following the labour reforms, the role of unions changed considerably. Prior to 1991, unions had monopoly bargaining rights. From 1991 onwards, union representation became an individual choice. The effect of the change was to reduce union membership by approximately 40 percent. In 1991, 40 percent of the paid workforce was unionized; just over three years later only 25 percent was. [Tim Maloney, "Has New Zealand's Employment Contracts Act Increased Employment and Reduced Wages?" Paper presented at The Fraser Institute's Right-to-Work Conference, Toronto, June 21, 1996]   Along with the reduction in union membership came a decline in the coverage of collective bargaining. Over the same period, the proportion of employees covered by collective bargaining was halved, to only 40 percent of all employees. [OECD, OECD Economic Surveys: New Zealand, 1996, p. 55] At the same time, the proportion of employees covered by individual contracts increased from 10 percent to 45 percent. [Lewis Evans, Arthur Grimes, Bryce Wilkinson, and Davic Teece, "Economic Reform in New Zealand 1984-95: The Pursuit of Efficiency," forthcoming in the Journal of Economic Literature.]

Assessing the immediate economic effects of the ECA is complicated because it is difficult to disentangle the impacts of the broader economic reforms that took place prior to the ECA being introduced. Nevertheless, there is an increasing body of evidence showing that the ECA has resulted in greater labour market flexibility and increased productivity. [OECD, OECD Economic Surveys: New Zealand, 1996, p. 56] One indication that today's labour market is more responsive is the widespread increase in the use of performance related pay. Further evidence comes from a comprehensive 1993 survey that asked employers to assess the effect the ECA had on number of different aspects of employment. The results confer an overall improvement in economic conditions for businesses. Some notable points include

•a net increase in employment

•nearly one half of respondents indicated that labour productivity increased, and

•63 percent reported an increase in the flexibility in operations.

Details of the survey are summarized in table 7.


As in Britain, the most important and identifiable benefit of labour market reform has been the improvement in employment prospects. The most recent economic recovery in New Zealand has generated an abundance of new jobs and an accompanying increase in the participation rate. The net impact has been a reduction in the unemployment rate as the employment growth rate outstripped the rise in the participation rate. Following a short period of adjustment after the ECA was implemented, job creation has been strong at 4.3, 4.7, and nearly 5 percent annually between 1993 and 1995 respectively. [OECD, OECD Economic Surveys: New Zealand, 1996, p. 9]   Canada's performance over the same years was 1.4, 2.1, and 1.6 percent. When New Zealand's ECA was first introduced, the unemployment rate was 10.3 percent. At the end of 1996, only 5.9 percent of the labour force reported they were unable to secure work. [Last quarter of 1996 is the most recent figure available, OECD, Main Economic Indicators, July 1997]

The benefits of labour reform in both Britain and New Zealand are indisputable. In both countries, structural labour market rigidities have been dramatically reduced, largely through the protection of an individual's right to freely negotiate his or her terms of employment. The objective in both countries was to make the labour market more responsive to rapidly changing economic conditions. Similar reforms in Canada would improve labour market operations and help reduce the unemployment rate.


Industries Regulated by the Federal Labour Code

The federal government has jurisdiction over industries that are both national in nature, and linked to all sectors of the economy. Federal labour laws govern bus operations, trucking, shipping, air transportation, and airports and related enterprises. In addition, the federal labour code applies to the telecommunications industry (including broadcasting and telecommunications), banks, postal services, and industries declared to be for the general advantage of Canada, such as grain handling and uranium mining. [Andrew Sims, Rodrigue Blouin, and Paula Knopf, Seeking a Balance: Canada Labour Code-Part I Review, p. 16] In total, the Canada Labour Code and its proposed amendments are estimated to apply to approximately 680,000 employees, or slightly more than 6 percent of all workers in Canada.[ Sims, Blouin, and Knopf, Seeking a Balance, p. 17] Liberalizing labour markets in these sectors, would create additional employment and an accompanying rise in productivity.

In addition to increasing employment within the federally regulated industries, liberalizing reforms would also bring about widespread secondary benefits. For example, cost reductions or increases in productivity in the transportation sector would have substantial additional benefits for almost all other sectors of the economy. As a trading nation, Canada depends heavily upon international and interprovincial exports. International exports alone account for 19 percent of all business in Canada. [Statistics Canada, The Economic Benefits of Interprovincial Trade in Canada, cat. 15-514, 1996, p. 18. This analysis and measure is based on international and interprovincial trade flows and is augmented by an integrated input-output accounting structure, thus allowing the linkages between production and exports to be identified. The linkages were used to "measure the economic importance of trade by estimating levels of gross domestic product and employment generated by exports, as well as examining industries producing goods and services that feed into the export chain] When interprovincial activity is factored in, the contribution of trade to GDP doubles. In 1990, total exports contributed $181 billion to GDP, or 37.5 percent of business activity in Canada. [The Economic Benefits of Interprovincial Trade, p. 18]   This production of exported goods and services also generated employment for 3.3 million people, or 1 in 3 private sector jobs. [The Economic Benefits of Interprovincial Trade, p. 18] All of this export activity is transported in some manner and thus would benefit from any efficiencies, cost savings, or reduction in industrial disputes generated by reformed labour legislation. Secondary benefits would also be generated throughout the economy through increased productivity in the communications industry, as all business in Canada depends on communications to some extent.

The simple fact that 94 percent of Canadian workers are covered by provincial legislation means that ultimately provinces would also have to adopt similar legislation to fully realize the potential reduction that comprehensive labour market reform would have on the national unemployment rate. Nevertheless, reform of the Canadian Labour Code is a starting point and an essential part of a broader strategy to address the unemployment crisis. That the industries are, by their very nature, of national strategic importance means that incorporating right-to-work legislation into the Canada Labour Code would have widespread benefits. Moreover, the provinces may be more inclined to pass similar legislation at the regional level once such legislation has been implemented by the national government.


Reforms in the Context of Federally Regulated Industries

"Liberals believe that a federal government must work with Canadian business to provide the proper supports and create a positive climate for economic growth." -Creating Opportunity: A Liberal Plan For Canada

To be sure, union powers and immunities in Canada are not nearly as extreme as they once were in either the UK or New Zealand. Nevertheless, the industrial relations environment does not need to degenerate to these extremes before it harms Canadian labour markets. There is an abundance of evidence showing that unions tend to reduce employment through reduced employment flexibility, higher wages, lower investment, and lower profits. Any legislation that enhances union security and aids the certification process will contribute to higher rates of unemployment.

By ignoring this reality, Ottawa's Bill C-66 moves in the wrong direction. Excessive union privileges in the UK and New Zealand led to high rates of unemployment and a relative decline in prosperity. To recover from the economic malaise, both countries underwent radical liberalizing reforms. Yet in spite of the evidence and experience of these other commonwealth countries, the federal government has drafted legislation that increases union security.

The federal government could generate highly effective reforms simply by protecting an individual's right to negotiate freely with an employer. The reform process would not need to be as long nor as comprehensive as in New Zealand or Britain. Instead, as part of a broader strategy to increase employment growth in the country, the Canada Labour Code could simply be amended to include:

•granting individuals the right to choose whether or not to associate with other employees for the purpose of advancing collective employment interests;

•giving individuals the right to join a union of their choice;

•protecting the individual's right to negotiate contracts with or without the assistance of a freely chosen agent;

•enhancing the protection of individual workers' rights and protection against wrongful dismissal; and

•requiring employers to seek individual written consent for the check-off of trade union dues.

While it is important to amend the federal labour code to protect a worker's right to negotiate freely, either collectively or individually, these changes alone will not be a panacea for the unemployment problem. Other factors contribute to unemployment such as employment insurance, minimum wages, etc. Further, under the Canadian constitution, labour is primarily a provincial responsibility. Therefore, true liberalization of all sectors of the labour market requires provincial adoption of similar legislation. Nevertheless, because the federal labour code covers key industries employing a significant number of workers, appropriate amendments to the federal labour code are an essential component of a strategy to lower unemployment in Canada.


Conclusion

The primary reason for this study is that there are 1.5 million people recorded as being unemployed in Canada. Indeed, a number much smaller than this should cause Canadians serious concern and prompt governments to respond with policies that would help solve the problem. However, as the preceding discussion indicates, appropriate policy changes have not been forthcoming. Rather than moving towards freer labour markets, the federal government is considering amending its labour code to produce additional rigidities within sectors of national economic importance. This will exacerbate rather than help solve the unemployment problem.

The magnitude and persistence of Canada's unemployment suggests the problem is largely structural. The natural rate hypothesis indicates that the solution lies partly in reforming the institutions and structures that determine the rate of unemployment. Thus, if governments wish to reduce unemployment, their efforts should be focused on the operations of the labour market.

Today, it is clear that the labour market reforms undertaken in the UK and New Zealand have been highly successful. These "real world" experiences should provide the federal government with sufficient evidence that liberalizing labour markets is one of the most effective strategies for permanently reducing the unemployment rate. In light of current fiscal constraints, such measures should be especially appealing as they do not entail foregoing government revenues.

This study does not suggest that reforming the Canada Labour Code is a panacea for the unemployment problem. Undoubtedly it is not, as more than 90 percent of all workers in Canada are covered by provincial labour laws. However, outlawing the forced remittance of union dues under the current Rand formula provision, and protecting an individual's right to bargain individually or collectively will result in a more efficient allocation of labour resources. Perhaps such reforms by the federal government would prompt a similar response by the different provincial legislators.


References

Addison, John and Stanley Siebert, (forthcoming). "Union Security in Britain," in Fazil Mihlar, ed., Unions and Right to Work Laws: The Global Evidence of Their Impact on Employment, Vancouver: The Fraser Institute.

Addison, John and Barry Hirsch (1989). Union Effects on Productivity, Profits, and Growth: Has the Long Run Arrived? Journal of Labour Economics, 7, 1: 72-105.

BlanchFlower, David and Richard Freeman (1992). Unionism in the United States and Other Advanced OECD Countries, Industrial Relations, 31, 1 (Winter).

Block, Walter and Michael Walker eds. (1982). Discrimination, Affirmative Action and Equal Opportunity, Vancouver: The Fraser Institute.

Brown, Charles and James Medoff (1978). Trade Unions and the Productive Process, Journal of Political Economy 86, (June): 355-78.

Buchanan, James M., Robert D. Tollison, and Gordon Tullock eds, (1980). Towards a Theory of a Rent Seeking Society, College Station: Texas University Press.

Canadian Chamber of Commerce, Submission to the Task Force-Review of the Canada Labour Code (Part 1).

Côté, Denise and Doug Hostland (1996). An Econometric Examination of the Trend Unemployment Rate in Canada. Bank of Canada Working Paper 96-7.

Clark, Kim (1984). Unionization and Firm Performance: The Impact on Profits, Growth, and Productivity. The American Economic Review 74, 5 (December): 893-919.

Daly, Donald J. and Donald C. MacCharles, On Real Wage Unemployment, Focus No. 18, Vancouver: The Fraser Institute, 1986.

De Matteo, Livio and Michael Shannon (1995). Payroll Taxation in Canada: An Overview. Canadian Business Economics 3, no. 4, (Summer): 5-22.

Denny, Kevin, and Steven Nickell (1992). Unions and Investment in British Industry. Economic Journal, 102, (July): 874-887.

Dicey, A.V., Law and the Public Opinion in England (1914). Macmillan, 2nd edition 1963.

Ernst and Young, (1995). B.C. Minimum Wage Study: Canadian Restaurant and Foodservices Association Final Report. Vancouver: Ernst and Young.

Evans, Lewis, Arthur Grimes, Bryce Wilkinson, and Davic Teece (forthcoming). Economic Reform in New Zealand 1984-95: The Pursuit of Efficiency, Journal of Economic Literature.

Freeman, Richard and James Medoff (1984). What Do Unions Do? New York: Basic Books.

Friedman, Milton (1968). The Role of Monetary Policy. American Economic Review 58(1), (March): 1-17.

Government of Canada, (1994). Agenda: Jobs and Growth, A New Framework for Economic Policy. Ottawa: Department of Finance.

Gregg, Paul, Stephen Machin, and David Metcalf (1993). Signals and Cycles: Productivity Growth and Changes in Union Status in British Companies. Economic Journal, 103, (July): 894-907.

Grubel, Herb (1997). Labour Market Rigidities and Canada's High Unemployment Rates. Paper presented at The Fraser Institute conference Right-to-Work Laws: The Global Evidence of Their Impact in Reducing Unemployment, Calgary, May 16.

Grubel, Herb and Joseph Bonnici (1986). Why Canada's Unemployment Rate is so High, Vancouver: The Fraser Institute.

Grubel, Herb and Michael Walker, eds. (1978). Unemployment Insurance: Global Evidence of its Effects on Unemployment, Vancouver: The Fraser Institute.

Hanson, Charles (forthcoming). The Economic Impact of Labour Reform: The UK Experience. In Fazil Mihlar, ed., Unions and Right to Work Laws: The Global Evidence of Their Impact on Employment, Vancouver: The Fraser Institute.

_____ (1991). The Taming of the Trade Unions. London: Macmillan Publishing.

_____ and Graham Mather (1988). Striking Out Strikes: Changing Employment Relations in the British Labour Market. Institute of Economic Affairs.

Henderson, David (1997). Canada's High Unemployment Rate Is No Mystery, The Wall Street Journal, February 7, 1997.

Hirsch, Barry (1991). Union Coverage and Profitability Among U.S. Firms, Review of Economics and Statistics 73, February 1991, pp. 66-77.

Hirsch, Barry (1997). Unionization & Economic Performance: Evidence on Productivity, Profits, Investment and Growth, Public Policy Source #3, Vancouver: The Fraser Institute, (June).

Holmes, Thomas (1995). The Effects of State Policies on the Location of Industry: Evidence From State Borders. Federal Reserve Bank of Minneapolis Research Department, Staff Report 205.

Kasper, Wolfgang (1996). Free to Work: The Liberalization of New Zealand's Labour Markets, Queensland: The Centre for Independent Studies.

Krugman, Paul (1994). Past and Prospective Causes of High Unemployment. Federal Reserve Bank of Kansas City Economic Review, (Fourth Quarter): 24-43.

Laporta, Pasquale and Alexander W. Jenkins (1996). Unionization and Profitability in the Canadian Manufacturing Sector. Industrial Relations, 51, 4: 756-776.

Long, Richard (1993). The Effect of Unionization on Employment Growth of Canadian Companies. Industrial and Labour Relations Review, 46, 4, (July): 691-703.

Maki, Dennis (forthcoming). Canada's Labour Laws: Where Do We Stand? In Fazil Mihlar, ed. Unions and Right to Work Laws: The Global Evidence of Their Impact on Employment. The Fraser Institute.

Mihlar, Fazil (1996). A New Labour Regime for B.C.: Towards Job Creation, Change and Choice, Vancouver: The Fraser Institute, February.

Noreau, Nathalie (1994). Involuntary part-timers. Perspectives on Labour and Income. Statistics Canada cat. 75-001E, (Autumn): 25-30.

OECD. Quarterly Labour Force Statistics.

OECD (1994). The OECD Jobs Study: Evidence and Explanations Parts I and II. Paris: OECD.

OECD (1995). OECD Economic Surveys: United Kingdom. Paris: OECD.

OECD (1996). OECD Economic Surveys: Canada. Paris: OECD.

OECD (July 1997). Main Economic Indicators. Paris: OECD.

Parker, Ron (1995). "Aspects of Economic Restructuring in Canada, 1989-1994." Bank of Canada Review. Ottawa: Bank of Canada.

Phelps, E.S. (1968). Money Wage Dynamics and Labor Market Equilibrium. Journal of Political Economy, 76, (August): 678-711.

Podgursky, Michael (1986). Unions, Establishment Size, and Intra-industry Threat Effects. Industrial and Labour Relations Review, 39, 2 (January): 277-284.

Ruback, Richard and Martin Zimmerman (1984). Unionization and Profitability: Evidence from the Capital Market. Journal of Political Economy, 92, 6 (December): 1134-1137.

Spiro, Peter S. (1996). The Effects of Fiscal Policy on Canadian Economic Growth and Employment, Ontario: Ministry of Finance, Working Paper.

Sims, Andrew, Rodrigue Blouin, and Paula Knopf (1995). Seeking a Balance: Canada Labour Code Part I Review, Ottawa: Public Works and Government Services.

Statistics Canada (1996). The Economic Benefits of Interprovincial Trade in Canada, cat. 15-514-XPE.

Statistics Canada. The Labour Force, cat. 71-001.

Statistics Canada. Labour Force Annual Averages, cat. 71-220.

info@fraserinstitute.ca You can contact us at the above email address for any comments or information requests. Please report any dead links or technical problems.

If you know someone who would be interested in this web page, please enter their email address below, and we will forward this URL to them: Email Address:

Last Modified: Wednesday, October 20, 1999.