instuto cultural minerva

institute of brazilian issues

the george washington university

Sandra Márcia Chagas Brandão

Washington, DC

Since the 1970s, the economic performance of the OECD countries has been very unstable. The two oil shocks and the sharp increase in interest rates in the 1980s created an unfavorable economic environment, interrupting the growth trend verified in the two previous decades. As a reaction to the new conditions, the majority of these economies adopted monetary and fiscal policies more restrictive, reducing even more of their rate of growth. In the beginning of the 1980s, it seemed that the virtuous circle of raising investment, productivity and real wages, that have been a major feature of the growth cycle after the II World War, was definitively concluded.

The effects of this new macroeconomic framework were almost immediately felt in the labor market, mainly through a rapid increase in the level of unemployment. Consequently, all the social welfare schemes that were organized during the "capitalist golden age" were put under pressure. The collective bargaining that has become an important mechanism to define wages and labor rights in the majority of these countries was considered an obstacle to adjust the labor market. Additionally, the different growth rates of wage and productivity began to be considered a major cause of difficulties in dealing with inflation.

The economic recovery and the expansion of employment in the United States after 1983 brought the European labor market institutions to the center of the debate. Contrasting the flexible and unregulated American labor market to the European system of social protection, many analysts and policymakers pointed to the different degree of flexibility existing in each labor market as the main reason for Europe’s employment problem (OECD, 1990).

There are two important questions to understand the centrality of the flexibility in this debate. First, which labor market practices did European countries have that made their labor market less suitable to deal with the new economic framework? Second, what exactly does labor market flexibility mean?

According to Blank and Freeman (1994), many different causal theories were formulated to explain how the European labor market performance was adversely affected by labor market institutions. The following relations were the most widespread established.

1. Legislation regulating the hiring and firing of workers, centralized wage settlements, and worker organizations inside firms were artificially strengthening the position of the workers. For this reason, European firms had their ability to adjust to the changing economic situation reduced.
2. The existence of unemployment insurance system and income support programs allowed workers to be more selective in accepting a job offer. These welfare mechanisms were slowing workers responses to new economic signals.
3. Due to restrictions imposed by legislation, the mobility of European workers was much lower than in American. This factor could explain mainly the great regional differences in unemployment rates.
4. Long-term job creation was being limited by excessively high wages. The reversion of the trend of increasing wage share was necessary to raise profitability, and to stimulate investment.

Based on these theories, labor market flexibility acquired a specific meaning in that debate. It was usually used to describe the speed of employment and wages adjustment to the new economic environment. Despite the fact that the notion of flexibility could have a wide range of applications, the limits imposed by institutional arrangements to wage and employment adjustments in face of economic shocks became the underlying justification for a more flexible labor market in European countries.

In this sense, the reforms necessary to improve the performance of European labor market seemed very clear. They should tighten provisions for unemployment insurance benefits, loosen dismissal and hiring regulations, decentralize collective bargaining, and increase labor mobility through training programs. Going beyond these recommendations, the focal point of the debate seemed to be the existence of a trade-off between social equity and economic efficiency in the labor market regulation.

Over the 1980s, many European countries implemented reforms in their system of social protection, in order to increase labor market flexibility (Table 1). These changes in labor market regulation were based on a widespread belief that the reduction of labor market rigidities would be at least part of the solutions for the problem of high unemployment and slow growth.

Sample Policies Changes Implemented over the 1980s to Increase Labor Market Flexibility in Europe

Source: Translated from Blank and Freeman (1994:29).

After almost two decades, it is possible to look at these processes and try to find some lessons of these reforms. Moreover, the favorable performance of American economy in the last six years and the fiscal and monetary restraints imposed by the monetary unification to the European countries created a completely different environment to analyze the costs of employment regulation.

Reviewing the experience of five developed countries – France, Germany, Italy, United Kingdom, and United States - since the 1980s, this paper intends to study the relation among their system of employment relations, the labor market development and the rate of economic growth. Confronting the reforms implemented in the labor market institutions with the evolution of economic and labor indicators, it will try to show that the nature of labor market institutions can not be a sufficient explanation for the different economic and employment performance.

Some hypotheses will guide this analysis. First, if the domestic and external conditions to enhance economic growth do not exist, even the most flexible system of labor regulation will not be sufficient to solve the employment problem. The firms decide to invest or produce considering alternatives ways of increasing their profitability. The perspectives of economic growth are essential factor to these investments decisions. Consequently, the existing employment level will be defined by the economic rate of growth.

Second, the institutional variables usually define the range of possibilities to firms or other economic agents to chose among alternative strategies. Therefore, given the way the products market and macroeconomic policies develop, the affects of economic growth on labor market will be mediated by the system of labor regulation.

Third, facing pressures placed by technological and economic changes, the institutional arrangements usually tend to change in a slower and less dramatic way. Most important, the reforms implemented tend to be conditioned by a complex array of interests and by the interactions among political, social and economic forces existing in the national environment.

The choice of the countries that would be object of this study rests on the intention of looking at very different experiences of building or changing the system of labor regulation. France is one of the best examples of complex legislative regulation, in which the government initiatives play a strong role in forging the actual patterns of relations between workers and firms. Germany has a tradition of cooperation between labor and firms within a high-institutionalized system of employment relations. Italy has a system characterized by a combination of collective bargaining arrangements made in an under-institutionalized framework, and a highly institutionalized implementation of social policies through an extensive indirect public intervention. United Kingdom was the leading country in the process of reforming the labor market, and it had done it through increasing the degree of legislative control on collective bargaining. Finally, United States has been, since the 1980s, the model of flexible and deregulated labor market, supposedly an example to the others countries. Besides, its system of labor relations has also been evolving in the last years, movement that deserves some attention for its peculiarities.

The study is developed in two chapters. The first chapter analyses some features of the system of labor regulation in these five countries, searching for their basic patterns and recent changes. As these systems are very complex and the reforms were made in many different levels, the study will not be an exhaustive one. On the contrary, it is focused on some special traits of these systems. More specifically, it tries to make a description of the patterns of collective bargaining, the role of each party in it, and the level of decentralization existing. It presents the basic characteristics of the legislation and procedures that limit the discretion of firms in terminating the employment relationship, and the trends in reforming them. To the countries where a relevant change was made, there is a description of the measures implemented to make the work time and the pay system more flexible. In addition, it shows the main aspects of the unemployment compensation schemes existing in each country.

The second chapter analyzes the evolution of labor market indicators in each country, comparing the current conditions with the one existing before and during the changes in their system of labor relations. Considering some basic economic indicators, this chapter also tries to show how close labor market and economic performance are intertwined, in the sense that the former is directly subordinated to the economic growth rate.

The study will be concluded with a summary of the major findings about the differences in the labor market performance and the trends of labor regulation in these five countries.

The institutions regulating employment relations have become a controversial issue since the mid-1970s. The great discrepancies in the labor market performance among major industrialized countries during the last two decades have been related to the different degree of employment regulation existing among them. Contrasting the U.S and European employment regulation systems, the academic and political debate had stressed the importance of a less regulated labor market to enhance productivity, to improve technological changes and to accelerated the economic growth. Because of massive complaints from the business community and under political pressure to reduce unemployment, many European countries took, during the 1980s, legislative steps to increase the flexibility of their labor market. These processes of changing the employment regulation are the objects of this chapter, which intends to analyze the experience of five industrialized countries: France, Germany, Italy, United Kingdom and United States.

The analysis will try to identify the basic features of each employment regulation system and the changes they were submitted since the yearly 80s. Five great topics will be discussed. First, the importance of each agent in the process of defining the rules. Second, the employment protection regulations existing, that consists of both individual protections against unfair dismissal and special procedures relating to collective or mass redundancies. Third, the development of legislation to define the scope of alternative flexible forms of employment. Fourth, the existence of rules concerned to working time and compensation flexibility. Fifth, the characteristics of existing system of unemployment compensation.

The French employment relations system has been undergoing substantial changes since the beginning of the last decade. Two aspects of this process must be emphasized. First, it was done without fundamental rupture with the political role of the unions and state in defining the levels of employment protection. Second, as a reaction to economic and political instability, it had an erratic pattern of development, sometimes combining steps ahead and back on the same regulation.

One distinctive feature of French system of employment regulation is the fact that the unions’ legitimacy rests, in last instance, on their existence as institution. By government act, they were given, at the end of World War II, the power to negotiate employment conditions for all French employees. For this reason, their capacity to engage in collective bargaining is relatively independent of membership mandate, as they are recognized as political institutions representing the collective interests of employees.

This special political arrangement has two implications. First, there is a prohibition of limiting the gains of collective bargaining to union members, and compulsory membership is illegal. Second, union pluralism is a natural consequence of this institutional role, as the plurality is thought to be a warranty to the freedom of expression and association.

However, this peculiar institutional form can be seen as one of the explanations for the chronic lack of members that characterized French unions. More important, it also explains why the union activists are more interested in building political projects within the confederations, instead of taking part in the collective construction of the representation of interests. As a result of these two aspects, "unions have also been able to participate actively in the negotiation of gains to employees without having to endorse compromises, since agreements come into force as soon as one of the representative unions signs them" (Saglio, 1995:205).

In the face of this political and particular role of French union, the extensive government intervention in employment relations is an expected result. The government is responsible for insuring the stability of the agreements between unions and employers, translating them into law and assuring their extension to every work. It should encourage, coordinate and arbitrate negotiations, and limit the agreement’s content to make them compatible with its policies. It can also propose and implement legislative and regulatory measures about any topic on employment relations. Besides these functions, the government is still the biggest employer in France, which gives it a strong instrument to shape benefits or define wage policy.

The collective bargaining process derived from these roles can be analyzed in two ways. From a legal perspective, it is conducted through a series of interrelated levels, with general rules defined at the higher levels. Negotiations at the shop floor level can only improve upon rules fixed at the company level, which must comply with the decisions taken at industry level agreements. Despite this hierarchical level of defining rules, the decision about which level is the appropriate to negotiate is subjected to the actor’s strategy, depending upon their logic or power.

From a perspective of its results, the negotiated agreement can be described as a declaration of a minimum standard level of protection or benefits that should be given to the employees. In this sense, it is not an instrument to predict the actual behavior of any employer, as the law allows them to depart from it. Moreover, as the implementation of the agreements is not based on strong grievance procedures, the possibility of disrespecting them is relatively high. For this, the French collective agreements have the basic role of providing "the standard for actors to refer to in assessing each other’s behavior" (Saglio, 1995: 210).

Considering these aspects of French collective bargaining process, it can be said that the statutory regulations has had, in France, a privileged position in labor relations, while the rules coming from the collective bargaining had a secondary role in defining the system of employment regulation. For this reason, during the 1970s, the legislation was built in order to achieve two purposes: a legal standardization of employment and working conditions, and a stabilization of employment relationships.

The legal changes implemented in employment regulations in France since the beginning of the 1980s affected this framework. To increase the flexibility of labor utilization, they aimed to reduce the direct restraint on management and to strength the role of collective bargaining in implementing labor regulation. Yet, they not affected the fundamental balance between capital and labor, which has been traditionally the core purpose of labor regulation in France and had justified the strong government intervention in the system of employment relations (Lyon-Caen, 1993).

The first wave of changing labor regulation in France during the 80s is known as Arroux Law, formulated between 1981 and 1983. The underlying purpose of these legislative actions was to strength workers’ rights in order to "transform the existing ‘code of labor relations’ into a ‘code of economic democracy’ (...), as a road toward liberating idle resources that could not be mobilized within a rigid setting of persistent asymmetries between capital and labor, unbalanced disciplinary powers in the hands of employers, and narrow job demarcations" (Lyon-Caen, 1993:348).

Among others, the following changes must be emphasized:

• increased the workers’ protection against employers’ abuses of their disciplinary power and strengthened workers’ decision capacity concerning work conditions;
• established equal opportunities for men and women in the workplace;
• promoted collective bargaining, establishing legal obligations to bargain at firm and industry levels. It had a special meaning to the process of wage determination, as firms of a certain size became obliged to hold annual wage negotiations, which gave legal basis to the firms define job classifications and pay differentials that do not stick to the provisions of industry level agreement;
• increased the scope for collectively bargained standards. From 1982 onward, the legislation recognized the possibility of negotiating terms that deviate from legal standards even in a way that was not favorable for employees. This increasing freedom for firm level agreements had great importance to negotiations about working time (duration and schedules), area that was the subject of many flexible agreements since 1982.

Another important subject affected by legislation changes implemented in the first half of the 1980s was the fixed-term contracts and temporary employment. The presidential ordinances of February, 1982, aimed to make the use of temporary work more difficult, defining that:

• temporary staff could be used only in two circumstances: replacement of an absent employee, and an exceptional or transitory work load;
• fixed-term contracts or hiring agency workers could not last for more than six months;
• temporary employees should have the same social rights in terms of salaries, fringe benefits and working conditions as workers hired on a permanent contract.

The second wave of changing French employment legislation occurred in 1985-86 and the several reforms had as their basic aim to increase labor flexibility and to adapt legislation to economic needs. Three major areas were affected by these reforms: dismissal of workers on economic grounds, restrictions on temporary work, and regulation of working time.

The first reform was implemented in many steps. In July 1986, it was made unnecessary official permission for layoffs when work force reduction involved less than ten works. In October 1986, the national interindustry agreement asked for the creation of retraining contracts for large-scale redundancies. Finally, in December 1986, there was a complete reform of the legal provisions to dismissal for economic grounds, abolishing all requirements to obtain official permission before dismissing staff. Also, the procedures were made much simpler and they must be completed, when prior consultations with the work council were involved, within a maximum period of thirty days.

Besides being an important measure to remove obstacles to quantitative work force adjustment, this reform stimulated a reorientation of legislative action and the emergence of regulations concerning qualitative work force adjustment. According to Lyon-Caen (1993), in the end of the 1980s, the government implemented many measures in this area, as the establishment of workers’ right to be financed to acquire new skills in the case of redundancies; a legal obligation on firms to provide a adaptation plan when promoting technological reorganizations; and the financial support of public agencies to firms that want to prevent work force reduction. "In other words, we encounter a new tendency toward replacing the hitherto prevailing model of quantitative employment stability with a model that emphasizes qualitative adjustments through labor mobility and human capital development" (p. 356).

The 1986 law on temporary work represented a turning point to the previous 1982 law about the same subject. It abolished the list limiting the circumstances in which the use of temporary work was permissible. It created a new type of seasonal employment contract, the intermittent employment contract, that is an indefinite work contract containing provisions for temporary interruptions. It extended the maximum period for temporary work arrangements to two years. Nevertheless, all the social benefits defined in the 1982 law were maintained.

The legislation regarding working hours was passed in June, 1987, and it had the purpose of codifying in a legal way many practices that were being adopted since the 1982. The law made possible variations in weekly hours when occurred simultaneously a reduction in average weekly hours and a company or sector agreement was signed. This legislation created legal background to mechanisms like the extension of the reference period for calculating average weekly working time, the individualization of working time schedules, and the organization of special weekend shifts.

Two other major changes in labor regulation were made since then. The first was done in July 1990, and was another legal reform of atypical employment. The new law reduces the maximum duration for the use of temporary work to twelve to eighteen months. It also limited the use of temporary work to circumstances such as temporary workload, seasonal work, or anticipated total abolition of the job. Despite appearing more restrictive than the 1986 regulations, this new law did not reduce the possibility of relatively free use of temporary work arrangements.

At the end of 1993, the Five-year Employment Act was approved to encourage the trend of agreements between firms and employees reducing both pay and working time in order to save jobs. Besides encouraging annualization and reduction of working time and amending rules applicable to part-time work and overtime, this law intended to grant state assistance in situations of reducing or re-arranging working time. The state is allowed to intervene in six procedures: compensation for short-time working; compensation for protracted short-time working; reduction of social security contributions payable on part-time jobs; assistance in transfers to part-time working; assistance in implementing reduced or annualized working time; and phased early retirement.

It is useful to make some comments about the French unemployment compensation scheme. As can be seen in Table 2, it involves both unemployment insurance and unemployment assistance, but with a strong insurance principle in terms of its financing.

Since mid-1970s, the regulation of this system has changed many times, in response to increase in the unemployment and to rise in the budget deficits. The trends were to increase the qualifying period and to link the duration of benefits more closed to the length of insured employment. Among the European countries analyzed, this later principle is most strong in France than in all other countries, except for Germany.

Trying to find the basic features of the process of changing the employment regulation in France during the last decades, it can be said that "the French way to flexibility is through expanding regulation" (Lyon-Caen, 1993:352). The search for a legal framework more suitable to economic and technological requirements resulted in the proliferation of state legislation, strengthening the major role of the state in labor regulation.

The legislative measures taken by the government can be understood as instruments to broad the scope for collective bargaining and to establish minimum standards and procedures for the implementation of alternatives ways of hiring and firing workers. Most important, they reflect a specific trend of action, that is a trend of public authorities to look to the collective agreements in order to find substance and justification to legal reforms. As the social partners react according to the economic and political framework, which has been very unstable in the last decades, this process of reforming the employment laws has been adaptable, and erratic, besides consistent with the rationale of French social legislation.

Contrasting with the French system, the German system of employment relations is much more balanced. The state have an important influence on the relationship between capital and labor through the laws to regulate dismissal, the social insurance program and some regulatory functions under the Industrial Relations Regulation Act. However, the direct negotiations among unions, work councils and employers have a strong effect in shaping the German system, through a "cooperative regulation of interests and a partnership between the actors in order to minimize conflict" (Baethge and Wolf, 1995:231).

In Germany, the system of interest representation rests on a two-tier institutional framework. At the industry level, the collective bargaining involves the unions and employer associations, which pursue their interests through high structured, centralized and bureaucratic organizations. This bargain is legally guaranteed and the state does not interfere directly in the process. The German Federation of Unions units the sixteen autonomous industrial unions, represents the workers in the political arena, and takes part of the process of formulating the basic principles for social and economic policy. United under the organization called United Federation of German Employers, 47 employer federations negotiate wages agreements with the unions.

The other level of negotiation is compound of two channel of employee representation at the firm level, which are the basis of the German system of codetermination or co-decision: work councils at the plant level and employee participation at the firm’s supervisory board.

The work councils are a non-union system of employee representation at the enterprise and plant level that must be established in firms with more than five employees. Their members are chosen by the entire firms’ work force and they are in charge of maintaining labor peace within the plant. To achieve this goal, they have legal rights established in the Industrial Relations Regulation Act of 1952 and the Amended Works Constitution Act of 1972. Among others, theirs rights include consultation and information rights on economic matters, codetermination rights on social concerns and personnel issues, and veto rights over personnel issues (Table 3).

The work councils negotiate wages and work conditions at the firm level, being oriented towards firms and workers’ professional interests. Doing this, "works councils generally relieve the unions from having to represent special interest groups and from having to deal with problems related to developments in production technology and work organization (so-called qualitative issues)" (Baethge and Wolf, 1995:235).

The other channel of workers’ participation at the firm level is a representation on firms’ supervisory board. This board is in charge of appointing and dismissing the top management, defining their remuneration and advising on general company policy. There are three types of representation, the first one established in 1951 at the coal and steel industries, giving full parity representation to the workers. The version established in 1976 covers all firms with over 2,000 employees, and requires that one of the employees representative has managerial status and gives to the chairman a casting vote. The third model is the weakest. It was also established in 1976 to cover firms with between 500 and 2,000 employees, and stipulates that one third of the supervisory board should be employees’ representative.

These arrangements can be said to be the key to the institutional stability and even efficiency of German system of employment regulation. The firm-level agreements negotiated by the work council allow a high degree of flexibility to the implementation of the standards defined by the collective bargaining. The negotiations between work councils and company management are in accordance with the specific competitive situation of each firm, and created a suitable framework to discussions about flexible work hours, wage policies based on workers’ performance and enterprise decision-making process.

Even during the 1980s, when the pressure over the European systems of representation were very strong, the German institutional structure look as if it was capable of adapting to the changing situation and to the new problems that were arisen. However, despite this basic structure of negotiation had not been changed in its supports, there was a gradual shift of the bargaining process to the firm level. The attempts to deregulate and to increase the flexibility of employment gave to the work councils the essential role in the negotiations. Progressively, the provisions of the collective agreement lost their meaning as uniform or generalized regulations, and became minimal provisions that should be made suitable to the conditions at each firm. The rising number of the so-called opening clauses in the collective agreements is another example of the increasing role of work council in the process of interests representation.

There are two consequences of this new trend within the German system of representation. The first one is referred to the fact that the existence of the work councils and their acceptance as partners in the process of defining the actual working conditions within the firms seems to have given to this system strength enough to resist to the several challenges posed by the new competitive framework. The knowledge and experience in defining at the plant level many issues that are fundamental to allow rapid adjustment seem to give the firms the mechanisms to go through variations in the demand level without having to force a rupture in the institutional structure of representation.

Two examples of this process could help to understand the importance of this two-level representation system. The performance-related pay system existing in Germany have been under pressure due to the introduction of group work, which includes new performances regulations, and calls for negotiations about plant-level performance measures. These negotiations should accommodate the demands of employers and the compensation claims of employees. In an experiment different from others seen in another countries, this discussion has been happening between work councils and company management, and the unions have focused their claims in the introduction of a skill-based system of work evaluation and wage setting.

Another example could be found in the results of the 1984 strike over work hours. The employees got a reduction of the average workweek to 38.5 hours, but the employers were granted increased working time flexibility. However, the agreement between unions and employers transferred to the work councils the responsibility of reaching firm-level agreements with employers, which increased differentiation among firms’ work schedule, as the regulation of these schedules was responsive to firms’ specific needs.

Another result of this strike allows us to analyze the second consequence of the increase role of the work councils. The German regulation of strike requires the union to provide monetary strike support to the affected workers. Until 1986, there was also a possibility that workers affected by reduction in hours worked due to production shortage receive unemployment insurance benefit. In 1984, the unions organized strikes at the most important automobile suppliers in order to force the auto manufacturers to cut production due to parts shortages. Doing so, the unions did not have to provide support strike to the workers affected.

Because of the subsequent debate about the methods adopted during the strike, a new law was approved in 1986 – paragraph 116 of the Work Promotion Act -, removing that entitlement. Transferring all the costs of strike payments onto the unions, this new legislation undermined their capacity to engage in draw-out strike activity.

Despite being the only great change in the institutional structure of collective bargaining, the government capacity of passing this law can be seen as a signal that the unions role have been going through substantial change. It is increasing difficult to them to continue to be centers of organizational support and coordination for work councils activities. With the increasing differentiation among the employees, the loss of social solidarity, which has been a fundamental root of the system of codetermination, is a danger to their role in the system and, for this, for its continuity.

This overview of the German system of interests representation seems to show that "the switch in Germany to a new regime of flexible accumulation did not require any fundamental breaks in established working practices, labor relations and methods of labor deployment but merely an intensification of existing patterns" (Lane, 1994:196). The same thing can not be said about the legislation on employment security, which was submitted to an important change in 1985, when the Employment Promotion Act (EPA) was implemented. It brought new regulations in the area of part-time work and job sharing, employment protection, use of agency works and conclusion of fixed-term contracts.

Protection against unfair dismissal has a long tradition in Germany and it is regulated by the state. Agreements among unions, work councils and employers have played a subordinate role in the development of these regulations. An outline of this comprehensive system of law should highlight the following features:

• dismissals and layoffs, except for disciplinary reasons, require advance notification, and this period increase with seniority;
• the law covers all employees working for more than six months in firms with at least six employees;
• all dismissals and layoffs must be grounded in just cause, which could have three basis: incapacity or poor performance of the worker, misconduct by the worker, and economic reasons including operational requirements of the establishment;
• collective layoffs for economic reasons should follow any "social criteria" to select workers, such as seniority, personal economic situation, and individual reemployment prospects;
• in case of unjustified dismissal or violations of procedures rules, the worker can sue the firm for maintaining the employment relationship or for monetary compensation through a labor court;
• all dismissals or layoffs must be approved by firms’ work councils. In case of disapproval, the employment relationship continues until a labor court decides if the dismissal or layoff was justifiable;
• the work council is entitled to demand a negotiation of a social plan to minimize social and economic effects of a major work force reduction;
• collective layoffs exceeding 6 a 10 percent of the firms’ labor force require notification to the local labor office, which may delay the process for up to a month and demand a introduction of short time working in the meantime.

These general regulations are part of a highly elaborate system of dismissal restraint that was only marginally affected by the 1985 EPA. The new legislation changed this institutional structure in two minor points. It partly exempted newly established small enterprises from the obligation to prepare a social plan in case of major work force reduction or production reorganization. It also changed the method of calculation of the minimum number of employees required for the application of the procedures of dismissal law, by excluding apprentices and marginal workers, thus increasing the number of small establishments not covered by statutory dismissal protection.

As the effect of these changes over the degree of protection against unfair dismissal were low, one could consider that the deregulation proposed by the 1985 EPA was irrelevant, mainly when comparing with the deregulation measures taken in other European countries. Nevertheless, the new legislation on fixed-term employment had strong impacts on the level of employment protection in Germany.

According to the provisions of the EPA, from 1985 onwards it was not necessary to have a legitimate reason to finish a fixed-term contract up to eighteen months of duration. The maximum period for the use of agency workers was extended from three to six months. It allowed firms to prolong the probationary period for newly hired workers for a period longer than the one established in the previous legislation. It also give permission to the newly established enterprises employing no more than twenty workers to use fixed-term contracts with maximum duration of twenty-four months.

However, these more flexible rules to hire worker on a fixed-term basis affected "job security only indirectly, by encouraging fixed-term hires that do not require any notice on the fixed date of expiration" (Buechtemann, 1993:287). It means that, instead of relaxing procedures to fire workers as other European countries have done, Germany made an option to reduce employment protection in the hiring side. It certainly strengthened the seniority principle that is a pillar of its dismissal laws and of its unemployment compensation system.

During the 1980s, Germany promoted many changes in its unemployment compensation system, mainly increasing the relation between the wage earned in the previous work and the amount of benefit when entering the system. In the beginning of this decade, the German system was composed of an unemployment insurance, which was closely related to the contributions made in the previous employment, and an employment assistance scheme to protect the ones who do not have the right to the other scheme for having exhausted the maximum period or not qualifying to it (Table 2).

As summary, it can be said that, compared with the instability of the economic environment and the high speed of introduction of new processes of production and labor management, the employment regulation system in Germany has gone through less changes than it was expected. As Baethge and Wolf (1995) said, "despite these challenges the institutional continuity of the German system is striking, especially in comparison with other highly developed capitalistic economies" (p.230). Certainly, the German unification will pose challenges still more stringent to the system. However, as many authors have emphasized (Birk, 1993; Buttler and Walwei, 1993, Baethge and Wolf, 1995), its capacity for finding in itself solutions to reorganize the relations between workers and enterprises rests in a peculiar sense of social solidarity. Until now, it has protected the role of the unions and postponed a more radical change in the level of employment protection. An important question is for how long the economic conditions will allow the continuity of changing the system without promoting its rupture.

During the 1980s and in the early 1990s, the system of employment relations in Italy has been transformed by major reforms. Its institutional design in the beginning of the 1980s could be described as a combination of collective bargaining arrangements characterized by a low degree of institutionalization and a multiplicity of levels, and a highly institutionalized implementation of social policies and extensive indirect public intervention (Regalia and Regini, 1995). Through continuous attempts to make the labor market more flexible and the industrial practices more innovative, this system was transformed in two complementary ways. The 1980s were a period of legal innovation aimed to promote external and internal flexibility. The 1990s have been a phase of reconfiguring collective bargaining and rearranging the relations among social partners in the system of employment relations.

As occurred in the majority of industrialized countries, in Italy there was a process of decentralization of collective bargaining in the 1980s. This movement happened without requiring substantial laws changes, as one of the basic features of Italian system was an absence of legal framework regulating relations between capital and labor.

Because of this lack of special regulation of the existence and operation of unions and employers’ organizations, the system of employment relations acquired some peculiar characteristics. First, the system of representation is pluralist, diversified, and fragmented. Second, until 1990, there were no regulations on strike activities or provisions to public mediation of employment disputes. Third, collective bargaining is not regulated, and there is no obligation to parties to a labor dispute to negotiate. Fourth, the bargaining structure is also largely unregulated, with negotiations being done in three levels – the national inter-industry level, the industry level, the company or plant level – without any formal definition of their respective competence and procedures.

In an open system of representation like this one, the shift of the negotiations towards the work place could be done without major ruptures. In the 1980s, the importance of enterprise level in the negotiation of agreements increased very fast. Focusing on the introduction of new technologies and changes in work organization and working time, these firm- and local-level contracts became the most important source of innovative agreements between capital and labor. Their increased importance also affected the sequence of contract negotiations, as "during the 1980s many local unions negotiated their own contracts first with the national agreements emerging later as an aggregation and ratification of these prior local deals" (Locke and Baccaro, 1996: 294).

This trend toward a decentralization of the bargaining process helps to understand two other basic changes verified in Italian system of employment relations during the 1980s. The first is the introduction of many working time arrangements oriented to give more flexibility for the full exploitation of the new technological processes introduced in many firms. These experiments included the use of overtime, additional shifts, and work on Saturdays in exchange for extra vacations, and flexible annual schedule of working time. They emerged as employers’ initiatives, but the unions had an important role to their spread to all sectors of economy by not raising objections to these kind of arrangements, and by allowing to place the discussion about these practices at the center of industry-level and plant level negotiations.

The second important change is related to the compensation composition. Until the beginning of the 1980s, the salary in most Italian industries had four components: a countrywide cost-of-living allowance linked to the scala mobile, a minimum wage rate fixed by industrywide national collective agreements, a firm-specific set of bonuses and premiums defined by collective agreements at the plant level, and an individual component granted unilaterally by management or individually negotiated.

Since the mid-1980s, forms of compensation more flexible have become central issues of negotiations, and they were rapidly widespread. The reduction of the influence of scala mobile, the establishment of skill differentials in the collective bargaining, the reintroduction of schemes of performance-related pay, and the strength of merit pays systems, all of these trends resulted in the increasing of the proportion of pay that is not defined automatically at the national level agreements.

Besides giving more flexibility to employers to boost employee motivation and productivity, this process of changing the structure of compensation was not opposed by the unions. On the contrary, it was an outcome of negotiation, in which the unions tried to improve their capacity of representing a more heterogeneous work force, strongly interested in getting monetary incentives for their performance. Due to this overlap of interests, "the move toward a more flexible pay system was the product of both re-regulation process at the workplace (through collective agreements and some involvement of workers’ representatives in the day-to-day management of the schemes adopted) and increased managerial attention to the more skilled employees" (Regalia and Regini, 1995: 150).

One basic requirement to this change in compensation structure was the process of revision and, in 1992, the abolition of the scala mobile. This was an economywide wage indexation system that was introduced by an inter-industry collective agreement in 1945. It was first reformed in 1975, when it became a flat-rate indexation mechanism. Since then, it also guaranteed 100% of indexation to industrial wages.

From 1975 onwards, the scala mobile continuously gained weight in the determination of Italian wages, and in the beginning of the 1980s, it was the determinant of over 60% of annual wage increases. As this increased importance of the scala mobile caused problems for all parties involved in the negotiations, this system was reformed again in the 1983 tripartite agreement. Since 1984, the government defined the rate by which the wages would be indexed to the inflation, regardless of the actual rate of inflation. Finally, as one of the main features of a national agreement, the scala mobile was abolished in 1992, and the social partners "established periodic tripartite consultations (in May and September) which would link wages increases at the national level to the government’s macroeconomic goals as stated in its yearly budget" (Locke and Baccaro, 1996:298). It also defined that the national contract on wage clauses should be renewed every two years and that wage increases negotiated at the plant level should be related to productivity gains or performance improvements.

The Cassa Integrazione Guadagni (CIG), another essential mechanism of Italian employment system, was also submitted to important changes since the 1980s. Unique to Italy, the CIG was created in 1941, through a collective agreement to compensate workers on temporary work reduction, and its scope was progressively broadened and its economic significance changed during the following decades. Besides subsidizing temporary layoffs, in the beginning of the 1980s, the CIG has become a large system to support industrial restructuring process, which also had the functions of:

• "disguising dismissals to defuse industrial conflict;
• (...) absorbing social shocks resulting from both social conflict and economic as well technological disequilibrium, either inside or outside the firm;
• public subsidizing of corporate restructuring or reorganization, that is, enabling firms to ‘externalize’ the opportunity costs of labor retraining and to have the whole economic system pay for it" (Tronti, 1993: 398).

Having these functions, the CIG was a central mechanism to finance the restructuring process of the 1980s. Being a substitute for dismissals, it created an alternative to the firms that were under strict rules to fire workers. Instead of firing them, they could apply to CIG benefits, having the freedom to adjust their work force both in qualitative and quantitative way. For the workers, it provided a compensation for reduction of work hours, a financial assistance for reemployment, and was an instrument of income maintenance.

The CIG has two branches to make its interventions in the labor market (Table 4). The ordinary intervention is made when there is the need of promoting work force adjustment vis-à-vis cyclical variations in economic activity. This kind of intervention is based on employers’ contribution, the benefits are temporary and they can be easily obtained. The extraordinary intervention is designed to deal with major structural adjustments. It is state financed, and its benefits can be prolonged indefinitely. Both branches allow compensation to partial (short time working) or total (layoff) reduction in working hours.

In 1991, the CIG was profoundly reformed. In order to reduce its excessive costs, the contributions were increased, the experience rates were made stricter, benefits limitations were established, and new time and eligibility rules were defined. To eliminate its negative effects over labor mobility, the Mobility Fund was created to take over the CIG functions of maintaining income and financing re-reemployment. Despite the changes implemented (Table 5), the CIG continued to be a very important instrument to cope with the need of flexibility to the firms to adjust their labor force that compensate the relatively strict regulation of dismissal that still exist. According to Auer (1993), "the flexibility of the CIG obviously permitted work force adjustment despite dismissal protection" (p.416). Moreover, "in the absence of a proper unemployment insurance system, the CIG acts as a functional equivalent to unemployment benefits and thus fills an important gap in the Italian system of social protection" (p. 422).

Two others reforms implemented in the early 1990s deserve some attention for their effects on the Italian system of employment relations. The first is the passage of two laws regulating strike in public services and employment relations in the public sector. The second is the attempt to reconfigure the collective bargaining structure and process.

Despite the initial opposition of unions, that were contrary to legal interventionism in industrial relations, a law regulating strikes and guaranteeing the provision of essential services in case of strikes was approved in 1990. It combines legal and contractual instruments to regulate this matter. The list defining which basic services should be provided in case of strikes was delegated to collective bargaining, process in which only unions that had already adopted a self-regulating code of strikes could take part. The law established restrictions on length and timing of strikes and created a Commission of Experts that is in charged of mediating conflicts, assuring the provision of essential services and disciplining the unions.

The reform in labor relations in the public sector was implemented in 1992 and 1993, and its major objective was "to eliminate a web of clientelistic work practices which were rampant in the public sector" (Locke and Baccaro, 1996: 297). To do so, it strengthened the autonomy of public managers, whose unilateral decisions could constitute an alternative to collective bargaining. It also introduced techniques of human resource management used in the private sector. It created an autonomous agency to negotiate collective agreements, to try to insulate the negotiation process from political pressures.

Besides their effects on rationalizing employment relations in public sectors, these laws are important for representing a legal intervention in the process of collective bargaining, initiative rarely seen in the Italian system. It also can be seen as part of a wide movement to reconfigure the system of employment relations, formalized in the July 1992 agreement among unions, employers and government, that also promoted a structural change in the process of representation.

The 1992 agreement affected the representation system in two different ways. First, it modified the structure and timing of the national bargaining. It defined that, every four years, the clauses governing hiring and firing procedures, job classifications, and career trajectories should be negotiated. Clauses focusing strictly economic subjects, like wages, should be renewed every two years at the national level. The bargaining should occur at both the industry and company or territorial level, in the latter level every four years. Most important to the unions, the agreement decided that the bargaining process at the plant or territorial level could established procedures only on issues not already regulated at the national level.

Second, it established a formal structure for firm-level worker representation, the Rappresentanze Sindacali Unitaire (RSU). The RSU are organizational structures of the unions and, at the same time, represent all workers. Two-thirds of their members should be elected every two years and the other one-third is appointed by those unions that have signed the national industry contract. The RSU members are recognized by the major confederations as legitimate agents in the bargaining process at the company level.

Considering the informality and the low level of institutionalization that have characterized the collective bargaining arrangements in Italy, these two changes represent a major rupture with the previous institutional framework. Defining the scope, procedures and competence of each level of representation, the arrangement established a more formal environment to the bargaining process, with clear links between various collective bargaining levels, but preserving a high degree of flexibility to make the decisions suitable to the unstable economic situation. Being done through a tripartite arrangement, these changes reinforced the Italian tradition of looking for concerted action among the agents and, most important, recognized the importance of strengthening the unions’ capacity to represent a highly heterogeneous work force.

The British process of changing employment relations system is a paradigmatic case. It is the extreme example of deregulation among all industrialized countries. It was the first process implemented as a package of legislative initiatives to increase the flexibility of the labor market. In addition, it was a movement of deregulation that promoted a rupture with the previous situation.

The changes implemented in the British system of labor relations since the beginning of the 1980s can be divided in two groups. There were measures aimed to alter the nature of British collective bargaining, mainly through a drastic reduction in trade union power. There were also measures to change individual rights, through decreasing the level of individual protection.

To understand these changes, it is necessary to consider that, different from other countries, in the U.K. there was relatively little direct employment regulation. The majority of the regulatory mechanisms were provided by the process of collective bargaining. The role of legislation was to provide protective "floors" where the collective bargaining had little or no influence. Most important, unionism and collective bargaining, the main source of employment protection, were legalized by the creation of a series of negative "immunities", legal protections against the force of the common law doctrines.

This specific way of legalizing collective bargaining has two consequences. First, there is no "right" to organize, to bargain or to strike in the British law. The capacity of the unions to organize the workers and represent them in industrial actions is not defined in the law in a positive way; on the contrary, it depends on "liberties" given by legislation. Second, "the very notion of a collective contract (...) does not exist in Britain; collective agreements are ‘binding in honor only’ and acquire legal relevance only to the extent that their terms are explicitly or implicitly incorporated into the individual employment contracts of those covered" (Hyman, 1993:321).

The importance of this peculiar institutional arrangement is that, to change the level of employment protection was necessary to reduce the influence of collective bargaining and the role of unions in the process of negotiating rights to the workers. The most important initiatives were taken in this area. In other words, "the deregulation of labor market was approached through weakening the instruments of employment protection" (Towers, 1993:330), namely the unions’ capacity of bargaining.

From 1980 to 1993, the government implemented new legislation or changed the scope of the negative "immunities" in order to reduce the union’s power. The legislative measures affected unions’ ability to protect their members in five different fronts.

Since 1975, the British unions were given a unilateral statutory procedure to get recognition as workers’ representative in the bargaining process. In 1980, this procedure was abolished by the government, the employers got the right to refuse to recognize or to bargain with trade unions, and they could withdraw recognition previously granted. Therefore, there was a reduction in the number and importance of collective bargaining.

The attack on collective bargaining was reinforced by governmental initiatives to de-centralize bargaining. Through examples when dealing with its own employees, or through dismantling arrangements based on national pay bargaining, the Government encouraged a trend towards a more decentralized and individualistic system of negotiation.

Part of the strength of the British unions rested on the closed shop clause. Since the beginning of the 1980s, the statutory immunities that legally protected closed shop had been narrowed, but only in the end of the decade, the union membership as a condition to employment was abolished. The Employment Act of 1988 established the same protection from unfair dismissal to both union and non-union member and gave the right to the worker to resign from the union. This latter change was especially important because, by prohibiting unions to discipline dissenting members, it creates the possibility of challenging the union power from within.

Another important legislation change occurred in the industrial action regulation. Miller and Steele (1993) called the initiatives taken in this matter "enterprise confinement", as the legislation tried to concentrate the law in the unit of employment. In 1982, the immunities to take industrial actions were narrowed to exclude political as opposed to trade disputes. In 1984, immunity for industrial action became guaranteed only if there was majority ballots. In 1988, individuals were allowed to refuse to strike without risking disciplinary actions by the unions. In 1990, the immunities were narrowed again, and secondary action should be protected only when it was a consequence of peaceful primary picketing. This continuous process of reducing possibilities of legal industrial actions and of focusing it at the plant level had the net effect of confining "lawful activities of combinations of workers within the gates erected by their employer" (Miller and Steele, 1993:230). In this way, it reinforced the rupture between the bargaining arrangement and the unions.

From 1980 to 1993, the British government implemented a series of statutes aimed to redefine ballots and union organization. In the case of legislation on union organization, the statutes defined a group of standards that the unions must comply when organizing their activities. Presented as a mechanism to increase union democracy, these rules reduced the autonomy of the unions in many ways. They decreased the reasons that a union could use to exclude a person from it. They defined stricter rules on union discipline. They establish a greater scrutiny and control on unions’ financial affairs.

The rules on ballots also helped to make the unions’ activity more difficult. Within the new statutes, all industrial action must be preceded by postal ballots. The unions must give notice to the employer of their intention to ballot and its results. Since 1984, unions must ballot every tem years to retain their political funds. Additionally, all financial help given by Department of Employment to offset the costs of postal ballots would be eliminate by 1996.

The measures taken in the area of individual rights had deregulatory effects in two different aspects. They limited or removed some statutory individuals rights, affecting unfair dismissal, claims for equal pay, race and sex discrimination and some rules on healthy and safety. They changed the regulation of pay, mainly through the abolition of the wage councils.

Since 1980, the regulations on unfair dismissal were changed in two ways. First, the minimum period of continuous employment to present a claim was extended from two to six months. Second, the responsibility for the burden of proof was transfer to the dismissed employee. In the area of women’s rights, the maternity payments and maternity leave have been made more difficult, but some new rights – as the right to claim equal pay for work of equal value, same age limit to receive redundancy payments and the removal of restrictions on hours and on specific jobs – were established.

This contradictory movement of reducing some individual rights and establishing others finds its explanation in the need of the British government to adopt some social rules defined by the European Community. In some areas, like discouraging discrimination, and collective agreements following takeovers, the implementation of EC regulation meant an increase in the level of individual protection in U.K. Additionally, the continuing growth of the degree of intervention on individual rights can be understand as a consequence of government’s attempts to remove barriers to contract some groups of the labor force, as it was the case of the limits to hire women in some specific sectors.

The abolition of the wages councils was a long process that was completed only in 1993. Established in 1909, the wage councils had their regulatory powers progressively extended until 1986, when they were responsible for fixing basic wage rates, piecework and overtime rates, holiday pay and holiday entitlements. At that year, the Wages Act restricted the scope of regulation of the wage councils, that became responsible for setting a single minimum hourly and overtime rate to employees above twenty-one years old. Finally, in 1993, the 26 wages councils, which had stipulated minimum rates for 2.5 million workers, were abolished, and the employers got the right to offer new contracts of employment at the market wage rate, whatever it was its level.

Besides the abolition of the wage councils, another important legal initiative in the area of deregulating pay was taken in 1980, when the schedule 11 of the 1975 Employment Protection Act was repealed. This schedule had given legal provisions to the comparability principle, that provided that the wage rate for a job should be equal within the industry, and had allowed the unions to seek a legally enforceable award against employers that did not follow the principle. Consequently, there were two major changes in the payment system in U.K. during the 1980s. There was a growth in individual performance pay both to manual and non-manual workers. There was also a growth in the coverage of group performance-related systems, including profit sharing.

It must be mentioned that the deregulation of pay was part of a government’s rejection of continuous income policies and national bargaining process. In this sense, it made the process of decentralization of pay bargaining deeper, by giving freedom to the employers to negotiate directly with their workers the level and structure of payment, and strengthening the move from multi-establishment to single-establishment bargaining.

The British unemployment compensation system is composed of an unemployment insurance benefit and a guaranteed minimum income (Table 2). In this system, the social welfare principles prevail, as the emphasis is on a minimum level of income, the payments are financed from progressive rather than proportional contributions, and the benefit is flat rate rather than wage-related. Since the 1980s, the unemployment compensation system was changed in many ways, but the most important reforms were done in the qualifying periods, that were made longer for those quitting jobs voluntarily, and in the benefits to young unemployed, that could be reduced if they refused an offer from the Youth Training Scheme or left the program.

The deregulation process of the British labor market has been seen as an exemplar case, for the high level of flexibility it had created. However, the main factor in the restructuring of the British labor market is not to be find in dismantling legal clauses of employment protection. The reduction of the influence of collective bargaining is the major determinant of the new patterns of employment protection. This peculiarity explains why the deregulation process has been done through an increase in the level of government intervention, and had made necessary to change in a radical way the balance of power in the collective bargaining. Considering these distinctive features of the British case, the lesson to be drawn of it is the impossibility of taking it as a model to be followed. It rested on a such specific institutional and political conditions that it only reinforced a common feature of almost every process of deregulation, that is the importance of the cultural and political traditions to forge a movement to cope with the challenges that new technological and economic requirements pose on the labor market.

The American system of employment relations has been, to many countries, the model of labor market they want to achieve with their reforms on labor legislation. This vision is justified by the fact that the American employers have no obligation, contractual or otherwise, to provide employment security, which seems to give them the necessary degree of flexibility to adjust their labor force whenever necessary.

However, the American system of labor relations is much more complex, and it could not be well described only looking at the employment-at-will doctrine. On the contrary, despite its institutional foundations has been preserved in its basic aspects since the 1930s, this system has been submitted, in the last two decades, to fundamental and structural changes, that affected both the level of employment security and the role of legislation in defining the workers’ rights.

The National Labor Relations Act of 1935 established the basic framework for American labor relations. It recognized the rights of the workers to collectively negotiate, and it defined the collective bargaining as the central mechanism to regulate and resolve the conflicts between labor and management. Nevertheless, in accordance with the American political and social dislike of government intervention, the NLRA did not choose the precise form of the collective bargaining or the conditions of employment. Instead, it "endorsed a process by which the parties could shape their own substantive contract terms" (Kochan et alli, 1994:24).

Another basic feature of the system enacted by the NLRA was that the strategic decisions and shop floor actions should be prerogative of the management. Besides being coherent with the tradition of business’ unionism, this principle created a clear distinction between the role of unions and of management, the former having the right to negotiate, through the collective bargaining process, the impacts of decisions taken by the management on employment conditions.

The third main aspect of this system was the capacity of the unions to exert a strict job control on the plant level, through highly formalized contracts and sharply delineated jobs. Codifying rules governing the use of labor in the firm, the unions could negotiate rules that attach a specific wage to each job and establish seniority rules to control career income and to exert influence on layoffs and promotions.

The evolution of the system based on these principles created, until the 1970s, a pattern for employment relations even where unions had not won representation rights. However, since that decade, the changes in the economic environment and the employers’ search for greater flexibility have made the set of rules governing labor relations in U.S. less standardized and have raised new issues in the American system of employment regulation.

Between 1960 and 1980, there was a growth of a non-union system of employment relations, at the same time that the NLRA system was developing. Many factors can explain the emergence of this alternative system. The higher increase in the level of employment in white-collar jobs, and in service sector than in blue-collar jobs and industrial sector changed the balance from productive segments in which the unions had penetrated intensively to others that their action was more fragile. This factor was reinforced by the development of managerial strategies to avoid the unions at the workplace and the increase incentives from the new competitive conditions to establish new methods of organizing the work and the production within the firm.

In the beginning of the 1980s, the American system of labor relations was compound by two different sub-systems. As the firms in the non-unionized sectors seemed to have higher flexibility to adjust their internal process of allocating workers, define wages, and restructure their process of production and labor force than the firms in the unionized segment, the practices in the former segment turned to be the example to the supposed necessary changes in the model.

The erosion of union coverage represented by the growth of the non-unionized sector and the recession in the first years of the 1980s are key determinants of the changes in the collective bargaining throughout that decade. The basic goal of the firms was to introduce a higher level of flexibility in the management of labor force. It was done through the decentralization of bargaining structure, movement that took the form of an erosion of inter-company pattern of bargaining, and the process of issues’ resolution shift from national level to plant level. It involved efforts to increase the flow of information about the economic situation of the firm among its labor force, in order to change expectations about the bargaining outcomes and to gain acceptance to new patterns of work and pay arrangements.

It also required changes in two practices that are on the heart of the union power, the work organization rules and the compensation structure. In the area of work organization rules, the main innovations were done through broadening job classifications, increasing management discretion in the use and allocation of overtime, liberalizing subcontracting rights, and removing restrictions on movements across the jobs. The changes in pay structure and practices meant basically a move from the system of attaching wages to jobs to one in which wages are more closely tied to individual skill levels and to the performance of the firm or the local labor market.

One consequence of the concessions made in these two areas were a expansion in the bargaining agenda, usually as a compensation for the changes in the work organization and pay rules. New employment security arrangements were introduced in many firms, mainly to "protect workers against job losses due to decisions where management has some choice, such as outsourcing, the speed at which new technology is introduced, and other efforts to improve productivity. The new employment guarantees do not deal with displacement caused by a general falloff in demand" (Kochan et alli, 1994:120; emphasis in the original). Additionally, many forms of gains sharing were experimented by the firms, and a great number of firms adopted employee stock ownership plans as new compensation arrangements.

These changes in the bargaining structure and in the rules to work organization and pay arrangements, together with the existence of a great non-union system, represented an important rupture in the American system of employment system. They were the ways firms found to "shift their internal labor market systems away from the traditional system of narrow job definitions and rigid work rules and toward much more flexible arrangements. Indeed, the United States is an outliner with respect to both Europe and Japan in terms of rigidity of its shop-floor systems" (Osterman, 1993:230).

This search for internal flexibility was happening at the same time of two others movements, the relative erosion of employment-at-will doctrine and the increase in direct legislation on labor market. These process had the contradictory effect of establishing stricter procedures on labor relations.

In U.S., there is not a comprehensive legislation protecting employees against arbitrary dismissal. The employment-at-will is the main doctrine in this area, and it asserts that employers can dismiss their employees at any time for any reason under any circumstance. During the past decade, this law has changed considerably, through the action of three forces.

The first is the inclusion, in collective bargaining agreements between unions and employers, of provisions to limit the possibilities of discharging employees only for "just" cause. These provisions usually are enforced through arbitration procedures, and the agreements normally provide that an employee dismissed in violation of just cause provision is entitled to reinstatement and back pay.

The second cause of the relative erosion of the employment-at-will principle is the development of state legislation restricting the right of employers to dismiss their employees. In the end of the last decade, almost every state have implemented some kind of statutes, based on the new wrongful-termination doctrines. At the federal level, the major focus of limitations to the employment-at-will doctrine has been the legislation against discrimination, and the rise in the reverse discrimination litigation, both giving indirect legal grounds to lawsuits against dismissals.

The third source of restriction to the employment-at-will principle is the emergence of numerous lawsuits involving disputes on wrongful discharge, sometimes involving high costs of litigation at the courts. It can be said that it is "the courts fashioning common law, and perhaps even more important, the juries administering their own form of both due process and justice, that have changed legal developments in recent years. (...) The result has been an uneasy coexistence between two systems of dispute resolution in the organized and unorganized sectors of the economy, the former taking the form of the arbitration process and the latter consisting of litigation in the courts" (Gould, 1996:66).

Another substantial change in the American system of employment relations during the last decades is the increase in direct legislation, formulated mainly at federal level. Since 1970, six major legislative package were implemented, aiming to established some kind of regulation in working conditions or individual and collective rights. "These developments, continuing through the 1980s, moved the state into a collective bargaining vacuum caused by the labor union decline, and thus gave Congress and state legislatures jurisdiction by default" (Gould, 1996:58).

Among these regulations, two are specially relevant to the level of employment protection existing in the U.S. The Worker Adjustment and Retraining Notification Act (WARN) was implemented in 1988, and it requires that employers with one hundred or more employees to provide workers or their representatives and local government officials sixty-day’s written advance notice before a plant closing or mass layoffs. This latter situation exists when 33% or more of the firm’s work force will be affected. The enforcement of these requirement is to be made by federal district courts, when the employees, a union, or a local government fill a suit to determine if the firm is exempted or not to comply with the WARN requirements.

Despite its intention to facilitate the displaced workers’ search for new employment or training, the WARN legislation does not offer universal coverage. Besides excluding employers’ with less than one hundred employees, the statutes contains great number of exemptions: employers are not obliged to provide advance notice if they are actively looking for alternatives to avoid closing the firm; if the layoffs are caused by a natural disaster; and if the dislocation is a result of business circumstances that were not foreseeable at the time when notice would have been required. It is important to call the attention to the fact that, in 1996, thirteen states have established laws regulating plant closings or relocations, in many cases closing the loopholes existing in the federal legislation.

The Civil Rights Act of 1991 is another federal legislation that has great importance to the actual level of employment protection in U.S. It restores much of the law that was established by title VII of the Civil Rights of 1964. However, establishing that punitive and compensatory remedies for employment discrimination violations of the law are appropriate, it goes beyond the previous law. This legislation has its main impact on workplace regulation, as it can play a major role in the reduction of inequities that are rooted on discrimination.

The system of unemployment insurance program in U.S. deserves some comments for its peculiarities. In this country, there is no national unemployment program; eligibility and unemployment benefits are set at the state level. The common feature of these state programs is that they are financed by a payroll tax on workers and employers.

Considering the aspects presented above, the development of the American system of employment relations in the last decades has been characterized by two opposite forces. There was a clear movement of changing rules on working conditions and on payment system, in order to increase the flexibility at the workplace. At the same time, there was a trend of regulating the employment relations, through federal or state legislation.

Both processes are unexpected in the ordinary interpretation of American labor market as the ideal of flexible and deregulated employment relations. However, the most trick feature of the American employment system rests on the possibility of, in the absence of regulation, the employees claim against an unfair dismissal in the common law, which is a countervailing force to the employers’ discretion. Combined with the increase scope of regulation, this aspect requires that the statement about the American labor market as the best example of flexibility should be taken with more criticism.

The previous analyses shows that the evolution of national systems of employment regulations since the 1980s has been very different among the countries considered. Certainly the most radical change has happened in U.K, where the reduction of legal provisions that allowed the unions representation made the employment protection much more fragile than in previous decades. Taking completely different steps, Germany and Italy implemented legislation to make their labor market more flexible, but they did so basically taking advantage of peculiar features of their system of employment regulation, mainly the tradition of concerted efforts among the social actors. In France, despite the changes have been done through legislative initiatives, reinforcing a country trait in dealing with this matter, there were a substantial departure from the previous situation, which qualifies the French reform of employment regulation as one of the deepest. United States offers the most interesting experience, as it combines a trend to make the workplace relations more flexible with a movement of increasing regulation on employment protection, a completely unexpected process considering the other national processes.

Despite these different approaches, it seems to have an underlying determinant to all these processes. In the five countries, the search for new patterns to the system of employment relations was done according to the historical, political and cultural framework existing. Despite the objective of the changes be the same – to increase the flexibility in the labor market to cope with the new economic and technological requirements -, the peculiar way of doing it was defined by the national background.

Another important conclusion is related to the effects of the deregulation trend on the European countries. Contrary to the expectations created by the intense debate during the 1980s, the pressures for increase numerical flexibility have had a limited impact on the European systems of employment regulation. Except for the United Kingdom, the other European countries considered in this study have not adopted a policy of outright deregulation. Instead, they have promoted concerted measures to increase flexibility in the labor market, but the various forms of flexibility have usually emerged as a result of tripartite consultation.

Putting together these two points, it is possible to say that there are no trend towards a model of employment regulation. The unevenness of the processes of changing the regulations, reinforced by the peculiar development of this matter in the United States, allows any analyst to expect outcomes that are anything but uniform.

In a recent study, OECD (1997) calculated some measures of collective bargaining to the countries that are members of the organization. Working with four different measures – collective bargaining coverage, trade union density, degree of centralization, and degree of co-ordination –, OECD intended to describe the effects of the substantial changes in the institutional arrangements on the structure of collective bargaining.

There is no doubt that it is very difficult to describe institutional processes in a quantitative way, and any method always involves some simplifications. However, these measures can be useful to have a broad picture of the changes in the institutional arrangements analyzed in the previous chapter. Considering the five countries that are object of this study (Table 6), the evolution of the mentioned characteristics deserves some comments.

The percentage of workers belonging to unions decreased in all countries, movement that was much more intense in France and in United Kingdom than in the other countries. However, the bargaining coverage has not evolved in the same pattern, as it has increased a little in France and Germany, and it had presented a slight reduction in Italy. Only in United Kingdom and United States, this measure followed the same trend of union density. This difference shows that, in those three countries, the collective agreements may have been extended by statutory regulations to third parties or by the employers to non-union workers.

Except for Italy and United Kingdom, the degree of centralization and co-ordination of collective bargaining did not change substantially in the countries analyzed since 1980. In Italy, the increase in the degree of centralization and co-ordination can be understood because of the 1992 tripartite agreement previously discussed. In U.K., decentralization of collective bargaining and the lower level of co-ordination can be seen as measures of the success of the initiatives taken during the 1980s to reduce the influence of unions on the collective bargaining.

Taken together, these measures shows that the evolution of the systems of collective bargaining has stressed the differences between the Anglo-Saxons countries and the others countries. Considering the degree of centralization and co-ordination, United Kingdom had moved towards United States, as it was expected. However, the influence of the unions is still stronger in the former country than in the latter, as shown by the union density and the bargaining coverage. The three continental European countries did some important changes in their system of labor regulation, without substantial effects on the measures considered. This peculiarity reinforces a previous comment that the reforms in these countries were made through concerted initiatives, and that they did not promote an outright deregulation.

Collective Bargaining Characteristics

1980-1994

Source: OECD, Employment Outlook, 1997.

1. Percentage of workers belonging to trade unions.
2. Percentage of workers whose terms of employment are determined by collective agreement.
3. Locus of the formal structure of wages bargaining. Values range from 1 (decentralized bargaining) to 3 (centralized bargaining).
4. Degree of consensus between collective bargaining partners, including both unions and employer co-ordination. Values range from 1 (uncoordinated) to 3 (coordinated).

The features of the collective bargaining and their development since the 1980s brings back the question about the possible impacts of the institutional arrangements on labor market performance. Despite done in an uneven way, the reforms promoted in the employment regulation systems were important and changed the process of defining contracts, work conditions, and labor rights. However, did they promote a rupture in the trends that had already existed in the labor market evolution? Did they affect the basic variables in a direct way or were their effects subsumed under the influence of macroeconomic performance on the level of employment?

Analyzing the trends of some indicators, this chapter will try to build a synthetic portrait of the labor market evolution in the period 1980-96 in those five countries. Comparing the behavior of the most important variables in these years and in previous periods, it will look for changes in the indicators trends that could be related to the new institutional framework. As the economic performance is an important determinant of the labor market evolution, the first section will present a very brief description of the behavior of some macroeconomic indicators to these countries. The second and third sections will establish some relations between the economic and labor market variables and analyzes the most important features of these latter indicators during the last two decades.

Many studies have already shown that it is very difficult to get definitive conclusions about the relation among the institutional arrangements, and labor market and economic performances (Blank and Freeman, 1994; OECD, 1997; ILO, 1996). Depending on the variable chosen and the hypotheses made, the statistical methods and the regressions can find the most different degree of correlation among these elements. Considering this question, this chapter do not have the intention of achieving a conclusive answer; on the contrary, its intends only to organize data that reinforces the perspective of analysis that the interaction among these elements is much more complex than the direct one existing in the deregulation debate of the 1980s.

The period between the Second World War and the first oil crisis is usually described as the most favorable to the capitalist economies. The stable, continuous and high growth rate verified in those years created the economic conditions to the establishment of the welfare systems and to the reduction of inequalities among regions and people.

Since then, there were great changes in the productive, technological, fiscal and monetary conditions in all developed economy in Western Europe and in the United States. For this, it is very difficult that the same process would be reproduced again. Nonetheless, for its uniqueness, the experience of those years makes it the parameter to judge the actual performance of any developed economy.

Examining the growth rate in the five economies, it can be seen why that period was called the "golden years" of capitalism. The average annual change in GDP achieved in those countries in the period 1960-73, for instance, is above the level of almost every subsequent period (Table 7). Therefore, the first feature of the relevant period for our study (1980-96) is that the growth rate of the five economies has been lower than the one verified in the full employment period (Chart 1).

Average Annual Change in Gross Domestic Product

1960-1996

Source: OECD National Accounts, 1998.

Beyond lower than in the 1960-73 period, the growth rates verified in these countries after 1980 have been very unstable (Chart 1). This characteristic is certainly even more important than the level of growth itself, for its effects on the investment decisions and government budget. Increasing the uncertainty about the future, this great instability of the growth rates, combined with the increase importance of the international financial flows, creates a completely different framework for private investment and for the government intervention in the economy.

In the 1990s, this last characteristic presented a significant difference among the five countries. Since 1992, U.S. economy has been growing in a more stable pace than in previous years. The others economies, despite growing in the period 1994-96, continued presenting a relatively more unstable trend of growth (Chart 1).

Another important feature of the current period is the change in the government behavior. Having to manage budgetary deficits, the government has had less freedom to implement fiscal counter-cyclical policies as it has done in previous periods. Looking at the figures to the 1990s (Table 8), it is possible to see why financing the budget deficit has been one of the most important government’s tasks in those years. To the European countries comprised with the monetary unification, the reduction of the deficit has been an even more important limit to its intervention in the economy. It occurs because they had to have, by 1997, a deficit equal 3.0% or less of GDP to qualified as members of the new currency.

The trend in private investment rate also does not support an extraordinary performance of these economies (Table 9). Except in the United States and United Kingdom, there was a fall in investment rates since the 1980s. Despite being difficult to define if this fall was caused by specific restraints on capital accumulation (as the high level of interest rates) or was a natural result of the slow-down in the economy, the final consequence was a decline in the potential growth of these economies.

Central Government Financial Balances

Surplus (+) or Deficit (-) as a percentage of nominal GDP

Source: OECD Economic Outlook, 1997.

Trend in Private Investment Rate as a Percentage of the GDP

Source: OECD Economic Outlook, 1997.

(1) Enterprise sector investment only.

The productivity growth weakened markedly since the beginning of the 1970s (Table 10). The question here is that, to avoid inflationary pressures or reduction in the profits, the increase in wages should be lower than the growth rate of labor productivity. Accepting this relation, the weak rate of productivity means that the wage-bargain systems existing should change to became suitable to the new economic conditions.

What can we learn from all these information? First, it can be said that, compared with the period 1960-73, the five countries have been characterized by a poorer economic environment. The instability of growth rate, the constraints to the counter-cyclical government intervention, the relative fall in the private investment rate, and the slow-down in productivity growth created a less favorable framework to the process of job generation. Whatever was the institutional regulation of the labor market, these new economic conditions would put under pressures the systems of collective bargaining and social protection rooted in an environment that did not exist anymore.

Percentage Changes at Annual Rates of Productivity in the Business Sector

Source: OECD Economic Outlook, 1997.

(1) TFP growth is equal to a weighted average of the growth in labor and capital productivity. The sample-period averages for capital and labor shares are used as weights.

(2) To Germany and Italy, the latest available year was 1994. To France and United Kingdom, it was 1995.

Second, these new economic trends had different importance in each country. United States is certainly the country in which the reconstruction of the economic environment was the most successful until now. The American economy has experienced two growth periods since 1980, the last one still in course. Having the international financial currency, the United States has also been able to use budgetary and monetary policies to sustain activity during the economic slow-down, as its government had done in the beginning of this decade. Additionally, for the size of its domestic market, U.S. is much more insulated from international instability than the other countries.

In all these features, the situation of the continental European countries is almost opposite. Confronted with the rules to monetary unification, they have been obliged to adopt restrictive budgetary and monetary policies. Despite the positive effects of commercial integration, their economies are still much more susceptible to external shocks than the American economy. For its turn, the British economy has been presenting some signals of recovering its strength, having an evolution more similar to the American one than of the other European countries. However, for its economic size, for the great challenges posed for the necessity of restructuring its economy, and the instability of its economic growth, it is not yet clear how successful had been the process of finding another path of development to the British economy.

Turning the attention to the labor market performance during the period 1980-96, we can take the civilian employment-population ratio as a good indicator of how dynamic the process of generating jobs has been in a specific economy. This statistic shows the proportion of the working age population that is employed, and it is expected to increase with this population if the unemployment rate is to stay constant.

The evolution of this statistic in the countries we have been analyzing shows that their experience is less than satisfactory, except for the results in United States (Chart 2). In France and Italy, the employment-population ratio has been decreasing since the beginning of the 1980s, movement that indicates the insufficient capacity of those economies to generate jobs in the speed required to maintain the unemployment stable.

Civilian Employment-Population Ratios

1980-1996

Source: U.S. Department of Labor – Bureau of Labor Statistics, 1997.

Until the beginning of the 1990s, the German and British employment-population ratios presented similar behavior, varying around the level registered in 1980, without a clear positive or negative trend. From 1990 onward, the differences between these ratios increased. German ratio decreased continuously, while the British ratio fluctuated around the 1980s’ average (Table 11).

Only in the United States, there was a clear positive trend in the employment-population ratio. In this country, this statistic was, in any year of the 1990s, higher than the average of any previous period. This sharp contrast with the European experience shows that, despite all the modernization process, the American economy could manage to create jobs in a rate above the growth rate of working age population.

It is important to call the attention to two other aspects of the employment-population evolution in these countries. The first is that, except for the United States, in all other countries the current employment-population ratio is lower than the one existing until 1980. Looking at this situation, we can say that the decrease in the economic growth rate had a huge impact on the capacity of these economies to create jobs. This conclusion is specially important because this ratio is currently lower even than the average registered in the 1970s, when these economies were trying to cope with the effects of the oil crisis, and had not done any substantial reform in their system of labor market regulation.

Civilian Employment-Population Ratios

1960-1996

Source: U.S. Department of Labor – Bureau of Labor Statistics, 1997.

The second feature is related to the level of the employment-population ratio in the continental European countries. Less than half of the working age population was employed in these countries in the end of the period considered, with the Italian situation being the extreme case. Even considering the possibility that part of those not employed being already retired, process that was stimulated by the government of these countries as an alternative to unemployment, this low ratio points to an important consequence of the relatively incapacity of creating jobs. An increasing proportion of their population is now depending on payments made by the system of social protection, through the unemployment insurance or the public pension system. Both transfers’ mechanisms put on pressure the already precarious budget situation, creating an additional and not desirable linkage between the employment problem and the fiscal imbalance.

The trends of the total employment-population ratios in these countries hide an important aspect that should be carefully considered by any program of reforming the labor market. Between 1980 and 1996, there was a sharp decrease in the male employment ratio in all European countries, ranging from 16% in Germany to 10% in United Kingdom. In the United States, this movement was not so impressive, as the male employment ratio decrease only 1,6% in that period.

To the women, the job opportunities seemed to have been greater than to the men, as in all countries, except France, their employment ratio increased in the period 1980-96. However, the rate of increasing was very different in each country: in United States, it rose 17% between those years; in United Kingdom, 12%; in Germany, 6%; and in Italy, it varied only 2%.

Considering these movements of the employment-population ratios for men and women, we can say that, especially in the European countries, the reduction or relative stability of total employment-population ratios hides another important process, the displacement of men of the labor market. Probably in all countries, including United States, this process is explained by the reduction of manufacturing employment, which used to be predominantly male jobs.

However, the important question here is not the cause of the movement but the challenges it poses to the labor market policies. If those men were already in the labor market and were displaced by technological or organizational changes, it is possible that their skills are no longer the ones required by the firms. If this hypothesis is right, there is not only a insufficient rate of job creation, but also a mismatching problem in the labor market, in the sense that the labor force looking for job could not have the necessary skills to get a job. To solve this situation, it is necessary investment in education and training to augment the chances of these people to find a job. If existing, this problem will be only marginally affected by policies aiming to increase external flexibility of the labor market.

Anyone analyzing the evolution of the employment-population ratios could ask if the results were not distorted by an unusual growth of labor supply in the period. The labor force participation ratio can help us to explore this question, as this statistic measures the proportion of the working age population that is working or looking for work. In this sense, this ratio measures the actual size of the labor supply in the economy.

The evolution of the participation ratios in the five countries between 1980 and 1996 shows that there was no special pressure on the labor market posed by an extraordinary rate of growth of labor supply (Chart 3). On the contrary, to almost every country, the participation ratios were quite stable during the period 1980-96. Again, the behavior of the American statistic is striking: it presented a positive trend, growing from 63.8% in 1980 to 66.8% in 1996, which means that 27,000 thousands people were incorporated to the labor supply in this period.

Civilian Labor Force Participation Ratios

1980-1996

Source: U.S. Department of Labor – Bureau of Labor Statistics, 1997.

Taking a long-term view, the stability of the participation ratios in the continental European countries during the period 1980-96 turned to be a reduction of the labor force size when compared to the working age population. In France, Germany, and Italy, the participation ratios in the 1980s and the 1990s were smaller than they were, in average, in previous periods (Table 12). Only in United Kingdom, the relative stability of this statistics in the period 1980-96 is reaffirmed when the time reference is broadened.

Exactly like the employment ratios, the split of the participation ratios by sex shows a very different trend in this statistic for men and women. In all countries, the male participation ratio decrease in the period 1980-96, movement that was more intense in the European countries than in United States. For its turn, the female participation ratios presented huge increases in all countries, varying 15% in United States, 13% in Italy, 12% in United Kingdom, 9% in Germany, and 7% in France.

Civilian Labor Force Participation Ratios

1960-1996

Source: U.S. Department of Labor – Bureau of Labor Statistics, 1997.

Considering the participation ratios behavior, the effects of the evolution of the employment ratio would be expressed directly through changes in the level of the unemployment rate. In the case of the continental European countries, the relative stability of the participation ratios, confronted with the declining employment ratios, allows us to expect a sharp increase in the unemployment rate. In the British case, the result would be less clear, depending on the effective differences between the two ratios. In the American case, the increase in the employment ratio, despite the increase in the participation ratio, would raise the expectation that the unemployment rate had decreased or, at least, continued stable.

Chart 4 presents the evolution of the unemployment rate in the period 1960-93 compared to the 1960-73 average unemployment rate. This last statistic is taken as a parameter due to the fact that those years are considered a full employment period. For this, comparing the average unemployment rate in 1960-73 with the actual rates will allow us to measure how much this statistic’s trend has moved away from that "ideal" level.

Clearly, the gap between the actual and the 1960-73 unemployment rates has increased in the European countries. In France and Italy, this movement has even been strengthened in the 1990s. In United Kingdom and Germany, after having doubled in the 1970s, the actual unemployment rate has been fluctuating around the 1980s average. It is necessary to call the attention to the increase in the German unemployment rate in the last three years, movement that was related to the effects of the reunification.

Again, the American situation is completely different from the European one. In United States, despite little above the 1960-73 level, the unemployment rate had not presented a raising trend in the 1980s or the 1990s. On the contrary, this rate has fluctuated around that average, and in the 1990s it has almost returned to that level. The most surprising aspect of this evolution is that it was possible in spite of the increase in the labor supply, which stress the extraordinary capacity of creating jobs that have characterized the American economy in the last two decades.

Focusing the attention in the unemployment rates registered in the 1990s (Table 13), it can be seen that, except for the United States and United Kingdom in the last three years, there was an increasing trend in this statistic in the continental European countries. As we have seen, at that period, many reforms had already done, their basic goal being to reduce the level of regulation in the labor market. In addition, the participation ratios in those countries had not presented a raising trend. If the demand for jobs was not greater than usual and the restrictions to hire and fire workers had been reduced in these countries, why the unemployment had not presented the expected behavior?

Unemployment Rates

1960-1996

Source: U.S. Department of Labor – Bureau of Labor Statistics, 1997.

Certainly, someone could say that the high and increasing level of the unemployment rate in these countries shows that the reforms were not enough to reduce the labor market rigidities, and to create a favorable environment to the firms generate new jobs. Nonetheless, considering all information presented until now, it is preferable to say that the still existing rigidities in the product market, that make difficult to these economies reentering in a new trend of economic growth, are substantial obstacles to the reduction in the unemployment rates. In these circumstances, the reforms in the labor market, whatever was their capacity of promoting ruptures with the previous situation, would not restore the entrepreneurs’ expectations. They will not hire more workers if they do not expect a continuous trend of growth.

The evolution of the British labor market is, in this sense, a good analytical example. There is no much doubt that the reforms made in the British system of labor regulation were very deep, changing the level of employment protection and reducing the restrictions to dismissals. However, the unemployment rate has been, during the period 1980-96, always above the average level registered in the 1960s and 1970s. Moreover, it had presented a declining trend only in the periods 1988-90 and 1994-96.

The American experience also deserves a close look when discussing this point. During the 1980s, despite the high level of deregulation of its labor market, the unemployment rate remained above the levels achieved in the two previous decades. It was necessary seven years of continuous and high growth rate to reduce the unemployment rate to the level registered before the 1980s.

Another way of trying to measure the effects of labor regulations’ reforms on the labor market performance is looking for the emergence of non-standard employment contracts, and for changes in the workers’ stability on their jobs.

Lets consider, for instance, the incidence of part-time employment in those countries (Table 14). Except in United States, the proportion of total employment that was part-time has increased in all countries since the beginning of the 1980s. United Kingdom has the highest incidence of part-time employment, as almost one worker in each four use to work only half of the usual hours per week. It is also interesting to note that, for Italy, if we look at the proportion of people working usually less than 30 hours per week, instead of using the national definition of part-time, the figure raises from 7.8% in 1983 to 10.5% in 1996.

Even recognizing that this increase could be, in some degree, related to voluntary part-time employment, its importance is not to be denied. Mainly, because these countries have made changes in legislation or procedures relating to non-standard forms of work, creating or increasing the possibility of hiring workers through alternatives contracts.

Incidence of Part-time Employment in Total Employment

National Definitions

1983-1996

Source: OECD Employment Outlook, 1997.

(1) Data not available.

Another form of non-standard employment that was subject of the regulatory reforms in these countries was the temporary employment. Information available to the European countries shows that, except in France, there was no great increase in the proportion of wage employees that were hired through this kind of contract. Between 1983 and 1993, the proportion of temporary workers in Germany varied from 9.9% to 10.2%; in United Kingdom, from 5.5% to 5.7%; and in Italy, from 6.6% from 5.8%. In contrast to this quite stability, the percentage of temporary employment rose sharply in France, going from 3.3% to 10%, between those years.

Probably, the reasons for these differences could be found in the different degrees of advantages an employer have when hiring a temporary worker. In Italy, for instance, the possibility of using the CIG mechanism to regulate the firms’ labor force when a demand fluctuation occurs reduces the importance of these contracts in this country. In France, the use of temporary contracts, mainly to reduce youth unemployment, has been a policy adopted and stimulated by the government.

The restrictions to dismiss workers were considered one of the most important rigidities existing in European labor market (European Commission, 1995). Despite being very sensitive to the economic cycle, data on short-time turnover could help us to check if the reforms on the rules to dismiss workers had any effect on the trend of this statistic.

Comparing the turnover between the first and second year of an employment match in 1985 and in 1995, we find that, as a proportion of those fired in each year, this turnover slightly increased in Germany (from 25.0% to 27.2%), in United Kingdom (from 40.5% to 42.9%), and in United States (from 60.5% to 65.9%). Certainly, the difference in the level of this turnover in the American labor market and in the two European labor markets is still big. But the trend can be seen as a signal that, at least in Germany and United Kingdom, dismissals are now easier than before the reforms, even considering that in the former country the reforms were made on the procedures to hire, not to fire employees.

The most important reforms in labor market regulations in all countries were made on the process of defining wages. It was changed in two complementary ways. The wage bargaining was decentralized, with the negotiations at the plant-level acquiring increase importance in defining the level of wages, and clauses guaranteeing automatic indexation were abolished almost everywhere. At the same time, there was a trend of increasing the proportion of the wage tied to individual or firm performance.

A comparative analysis of the trend in real earnings can help to describe the other side of the labor market performance in a period of reforms. Looking at the evolution of the compensation per employee, it is possible to say that, as an average, the behavior of this statistic in the European countries was much better than in United States in the past ten years (Table 15).

This difference becomes even more striking when the recent development of the earnings of the full-time workers is considered. In United States, whatever is the time reference chosen, there was a decrease in the real earnings of these workers, despite this movement had been less intense in the 1990s.

The contrast with the European countries could not be stronger. Taking, for instance, the evolution of the earnings of full-time workers in the past five years, they increase in all European countries, at rates that varied from 9.9% in Germany to 0.8% in Italy. In United States, there was a slight decrease (0.9%) in wages in this period.

The recent wage development in United States puts this country in a much more adverse situation than when we look at the employment ratio or the unemployment rate evolution. These figures seem to suggest that the price to achieve its superior employment performance was a sluggish real wage growth.

Splitting the full time workers by sex, we can see that, in all countries, the women’s earnings had a better evolution than the men’s earnings. The only exception has happened in Italy in the past five years, when the increase in earnings was bigger to men than to the women. Again, the United States experience calls the attention: the earnings of the men working full time decreased in the past ten years likewise in the past five years.

Real Earnings (1) Growth for Different Groups of Workers over the Past Five and Ten Years

Source: OECD Employment Outlook, 1997.

(1) All nominal wages series have been deflated by each country’s consumer price index.

(2) Youth refer to 21-25 year-olds for France.

(3) Prime-age workers refer to workers aged 31-40 for France.

(4) Data not available.

Except in Germany, the earnings of the prime-aged (25-54 years) workers presented a better evolution than the wages of the youth workers. However, in France, the difference in the rate of increase in earnings to these two populations was not great. In the United States, the difference was not the size of the increase, but the size of the decrease in earnings of these two populations: the 20-24 year- old workers had a greater reduction in their real wage in the last ten years (and also in the past five years) than the prime-aged workers.

Finally, Table 15 also allows us to analyze an important feature of the wage development in the last ten years in many industrialized countries, which is a sharply rising in the wage inequality. Except in Germany, the increase (reduction) of the earnings of the high-paid workers was bigger (lower) than the variation observed in the real earnings of the low-paid workers. In the case of the United States, and also Italy in the past five years, a particularly disturbing feature of the rise in wage inequality is that the situation of low-paid workers had deteriorated not only in relative terms (as was the case in France and United Kingdom) but also in absolute terms.

The results of this process can be summarized through the ratio of high to medium wages. In 1989, this ratio was equal 1.99 in France, 1.64 in Germany, 1.44 in Italy, 1.83 in United Kingdom, and 1.97 in United States. In 1994, it was equal 1.99 in France, 1.61 in Germany, 1.60 in Italy, 1.86 in United Kingdom, and 2.01 in United States.

What can we conclude from all these figures? First, it is possible to say that, during almost all the period 1980-96, the macroeconomic environment in those five countries were less favorable to the process of generating jobs or even to sustain the systems of employment regulation than in previous decades. The only exception was the American economic framework since 1992, as it combines continuous growth, rising private investment and productivity for a longer period than the previous cycles.

Second, except for Germany, there was not a substantial change in the evolution of the employment-population ratios in the period 1980-96. The French and Italian ratios continued to decrease, in spite of any positive variation in the growth rate (Chart 5). The British ratio continued to oscillate around the 1980s’ average, without taking on any special trend. The American ratio increased during almost the period. The German ratio, after returning, in the end of the 1980s, to its level in the beginning of that decade, presented a sharp declining trend in the 1990s.

This relative independence between growth rate and employment-population ratio that characterized the European countries in this period is a striking feature. If we accepted that these countries made important reforms in the institutional framework of their labor market, the absence of significant changes in the global capacity of creating jobs, mainly in the European countries, stress a point previously presented, which is the complexity of the relation between regulatory systems and labor market performance. Without suitable macroeconomic conditions, the employment problem will not be solved only by rearranging the laws and procedures that regulate the employment relations.

Third, the increase in wage inequality represents an important rupture with the previous experience of these countries. This higher inequality is an additional problem to economies that have not been able to create enough jobs. At the same time, looking at the American experience, this trend in wage inequality poses the question if, in the new economic conditions, there will necessarily be a trade-off between the level of employment and the wage inequality. For its effects on living standards, this last problem deserves more attention, mainly because the reforms in the labor legislation tend to reinforce this trend.

Economista formada pela Universidade Federal de Minas Gerais, Sandra Brandão fez o mestrado e o doutorado (1994) na Universidade de Campinas, assumindo depois o cargo de analista senior da Diretoria Adjunta de Análise Socioeconômica, órgão da Fundação Sistema Estadual de Análise de Dados SEADE, fundação pública que está ligada à Secretaria de Economia e Planejamento do Governo do Estado de São Paulo. Sua área de atuação abrange pesquisas sobre condições de vida, inclusive elaboração de análises sobre a situação socioeconômica das famílias na Região Metropolitana de São Paulo, sob um enfoque multissetorial. É coordenadora do Projeto Educação e/ou Trabalho: a ótica do trabalhador, realizado em convênio entre a SEADE e a Finepe, do lado brasileiro, e o Banco Interamericano do Desenvolvimento, BID.