|lPolkyRoh 1 WRPS 1earc WORKING PAPERS Education and Employment Education and Social Policy Departmnent The World Bank February 1993 WPS 1099 Trade Unions and Collective Bargaining Harry C. Katz Sarosh Kuruvilla and Lowell Turner The ability of developing countries and the new transitional economies to compete in the global marketplace will depend on their ability to transform industrial relations policies involving trade unions and collective bargaining so that they promote flexibility in the workplace and encourage the fornation and effective use of human resources. ThcyRcsrcWoing Ppcdiabtethcfindugof woknpwgessand eaew pthecxchangcofideasamongBankstaff andaU lheintcrstcdindevelopmentiscs Tescpapet distbedby teRescah AdvysSutfcarythcenamcs oftheauthou. eflectdytheuview.andshldbeusedanditedwgly.Ther,din tionanduonsarethcauthnown.They should not bc attuNtd to the Wodd Bank, it Bourd o Dims, its managnmc or any of its mcmber coumtiies Policy Research Education and Employment WPS 1099 This paper - a product of the Education and Social Policy Department - is part of a series of state-of- the-art studies on employment and labor market issues and reform programs. Copies of this paper are available free from the World Bank, 1818 H Street NW, Washington DC 20433. Please contact ESP, room S6-035, extension 33680 (February 1993, 39 pages). Katz, Kuruvilla, and Turner assert that changing Recent pressures for structural change in the world markets and new technologies are driving developing world present an opportunity for industrial restructuring. The ability of developing major transitions in industrial relations. countries and the new transitional economies to compete in the global marketplace will depend Drawing on the Japanese and German on their ability to transform industrial relations experiences, as well as experiences in the policies involving trade unions and collective developing world, Katz, Kuruvilla, and Turner bargaining so that they promote flexibility in the focus on lessons that can be applied in guiding workplace and encourage the formation and this transformation. -ffective use of human resources. Worker participation in decisionmaking, History has shown, they say, that there are they contend, is critical for bringing about the certain key moments of transition in industrial essential popular acceptance of changes that will relations systems. After that time, systems get set come with industrial restructuring. and are hard to mcdify. Often these key moments are the result of legislative changes It is also important to coordinate and inte- (such as the National Labor Relations Act and grate industrial relations policy with other social, the emergence of public sector unions after the legal, economic, and educational policies. The burgeoning of public sector legislation ir- mae education system, for example, should not be United States). Sometimes they are the result of overproducing college graduates when there is key historical or economic junctures (such as the an undersupply of unskilled and skilled workers. postwar reconstruction in Japan and Germany, and independence movements in the developing world). The Policy Research Working PaperSeriesdisseminates theftndings of workunder way in theBank. Anobjectiveof the series is to get these findings out quickly, even if presentations are less than fully polished. The fndings, interpretations, and conclusions in these papers do not necessarily represent official Bank policy. Produced by the Policy Research Disseininatic a Center Trade Unions and Collective Bargaining Harry C. Katz, Sarosh Kuruvilla and Lowell Turner New York State School of Industrial and Labor Relations Cornell University This paper is prepared for the Education and Social Policy Department of the World Bank. It is part of a series of studies to improve labor policies for managing the social cost of economic adjustment. Table of Contents Introduction: The Economic Context ................. ................... 1 Industrial Relations Issues in Developing Countries ........................... 3 The Economic Context .3 Industrial Relations Characteristics .5 The Pressures on Industrial Relations .10 Basic Principles of Law and Institutions .12 The Need for Unions, the Right to Strike, and Colletive Bargaining. 1 2 Independent Mediation Entities .14 Proper Union Roles .15 Workplace Representation .16 Wage Setting and Bargaining Structure .18 Bargaining Structure .18 The Disadvantages of Fragmented Bargaining Structure .19 The Virtues of Coordinated Bargaining .19 The Virtues of Annual Pay Agreements .20 The Occupational Scope of Bargaining Structure .21 Recent Movement to More Decentralized Bargaining Structures .21 Workplace Issues ........... 22 Recent Restructuring in Industrial Relations .23 Contingent Compensation .23 Team Systems ..................24 Employee Involvement in Decision Making .25 Difficulties in Diffusing Participation . 25 Employment Security .28 The New Industrial Relations and the Third World .29 Trairning and Work Force Skill and Career Development .30 The Need for Greater Investment in Training .30 The German Training System .31 The Extensive Japanese On-the-Job Training .33 The Mismatch Between Social, Educational and Industrial Relations Policies in Developing Countries .33 Conclusions .35 References ....... ... ............................................ 37 A. Introductlon: The Economic Context We live in an era of industrial relations reform and transformation. This is true across a wide range of advanced, newAy Industrializing and developing countries. While thirty years ago it was commonplace among social scientists to assume that in the long run other countries would converge on variants of the 'most advanced" U.S. model of collective bargaining and industrial relations, this assumption has been swept aside by history. The rise and success of very different models in countries such as Japan and Germany has demonstrated both a cross-national diversity of stable industrial relations pattems and new lessons regarding what does and does not work. The purpose of this paper is to draw out, from both more and less successful recent cases, some contemporary lessons for industrial relations reform efforts, especially for developing countries, and for the transitional economies of Central and Eastern Europe. The important starting point is a recognition of the impact of changing world markets and new technologies in driving industrial restructuring and industrial relations transformation. Above all, perhaps, the rapid growth of Japanese economic strength, rooted in part in a successful system of enterprise unionism and shopfloor teamwork, has put pressure on firms in other countries to reorganize production and work and to seek some kind of unew industrial relations." Within Europe, not only the advent of Japanese competition but the relative success of German industry and industrial relations, along with the more recent 'relaunch" of European economic integration, has pressured British, French, Italian and Spanish (among other) firms to reorganize. Everywhere, intensified world market competition has called into question established relationships and ways of doing things, and called forth managerial imperatives to raise productivity and product quality, achieve new flexibility, and cut costs. At the same time, and closely related, the past fifteen years have witnessed a rapid spread of new microelectronic technologies in the workplace. "New production concepts,' often based on the new technology, aim to make more flexible use of labor, both in the products produced and the process. As managers demand more responsibility from (and sometimes even give "semi-autonomy" to) individual workers and groups of workers, traditional labor-management relationships are called into question. 2 So widespread are these changes In the modem workplace, especially in advanced societies, that many have declared the era of Fordism at its end. "Post-Fordisma has been variously defined and labeled as flexible specialization, flexible automation, lean production, and diversified quality production, to name a few. Critics have called i neo-Fordismm or management-by.stress," implying a more sophisticated form of mrnanaerial control over the workforce (Parker and Slaughter 1988). But however one interprets the new developments, there is general agreement that the modem workplace is undergoing major changes. Part and parcel of these changes is a widespread rethinking of established pattems of industrial relations. In some cases, employers have gone on the offensive against union influence that is perceived to defend rigidiies in the workplace and stand In the way of nAcessary work reorganization. In other cases, employers have sought a new collaborative relationship with entrenched unions; and sometimes the second strategy has followed the first. Unions, for their part, have in some cases fought against the changes and defended the traditional ways on which their power is based. But in other cases, unions have demonstrated a willingness to cooperate In new ways, both to save jobs and to promote the human-side benefits of more flexible work organization. And in some cases, where they have not been excluded, unions have played a proactive reformist role, pushing management toward new forms of organization such as group work. In addition, the state, through public policy, has often played an important role (through regulation, deregulation, or other policy changes) in promoting industrial relations reform. In the interaction of world mnarket changes, new technologies, and the strategies of governments, employers, and unions, there is, theretore, considerable fermentation and change in contemporary industrial relations in advanced societies. For different reasons, although not wholly unrelated, there is also pressure for change and in some cases full-fledged transformation within the newiy industrialized countries (NlC's), some developing countries, and the various countries of Eastem Europe and the former Soviet Union. In this period of uncertainty, trial and error, and change, there is also a great deal of cross-national, intertirm, and interunion communication and exchange. No one wants 3 to reinvent the wheel; less successful firms and countries attempt to leam the lessons of the more successful, especially Japan and Germany. Sorting out the positive and applicable lessons has become a major preoccupation for managers, and to a lesser extent for govemments and unions. The opportunity to leam and combine lessons from various versions of successful (if not 'bestT) practice is one of the benefits of Intense contemporary processes of economic globalization. Nowhere is this need greater than among developing countries and the emerging and unstable democracies of Eastem Europe and the former Soviet Union. We want to emphasize, however, at the outset: it is not possible to transplant another model, either discrete elements or the model in its entirety. In every case, it Is necessary to build on existing practices and institutions, or institutional remnants. How the lessons presented in this paper can be adapted for specific use in specific national settings Is a matter for careful case-by-case study and experimentation. B. Industrial Rellatons Issues In DeveloDina Countris The changing economic environment confronting third world countries makes it Imperativf, that they take immediate decisions regarding industrial restructuring. The Economic Context In the 1980s, the U.N. General Assembly set a target of 7 percent for third world growth. This target was based on estimations that a 7 percent growth rate was the bare minimum for developing economies to provide basic needs (food, shelter, clothing and education) of their populations. However, the record after 1960 has been disappointing, as most developing countries plunged into deep and continuing economic crises (Singh 1992). On average, GDP growth rates during the 1980s were well below 5 percent and a number of countries experienced substantial declines in national income. Although African economies grew at 3.5 percent between 1960-73, the rate of growth was 1.3 in 1981, and negative thereafter. Mexico's GDP grew at 7 percent between 1977-1981, but 1982-1988 saw negative growth. Estmates (after GDP figures are adjusted for terms of trade and net factor payments) indicate that the 4 average per capita Income in Latin American countries was lower by 15 percent and for African countries lower by 30 percent in 1989, relative to 1980. Real wages in Africa have declined by about 20 percent and in Mexico, by almost 50 rercent between 1982 and 1987. The inability of third world countries to compete in world manufacturinig has also resulted in deindustrialization in many countries (e.g., Tanzania, where manufacturing production shrank by 25 percent during the 1980s). These economic pressures have forced developing nations into increased indebtedness, resulting in an increased burden of debt servicing. In the 1980s interest payments for most Latin American and African countries comprised about 30 percent of the national budget, seriously affecting ability to import necessary capital and technology for production and exports. Sachs (1990) suggests that the collapse of living standards in the third world in the 1980s (he calls the 1980s the lost decade of economic development) is intimately related to the extemal debt crisis and shows that economic performance has been disastrous in countries that experienced debt service difficulties. In addition, there was in the 1980s a net outflow of capital from the third world, in spite of large transfer payments via aid into the third world. Exceptions to this bleak scenario are the experiences of some Asian countries (mainly the newly industrializing countries), which showed increases in GDP growth rates and per capita income, and achieved both ILO and U.N. targets. The three main features that appear to account for the dramatic relative economic success of the Asian nations are: economic structure, economic and industrialization policies, and lower indebtedness (thus insuring relative protection from changes in world interest rates) (Singh 1992). Experts by and large agree that both extemal factors (world interest rates, terms of trade that discriminate against developing countries) and flawed domestic colicies are important in explaining economic decline, although there is debate regarding the relative impact of both. There is, however, consensus that the central problem for these countries is that & nigh long-term rate of economic growth is an economic and social imperative. There is also consensus that developing nations are not likely to reach target growth rates of 7-8 percent without fundamental changes in the structure of their economies 5 and economic policies. Finally, there is consensus that these countries need to improve the utilization and management of their human resources. F'fective Industrial relations practices are needed to achieve those ends. The key aims of an industrial relations policy should be to foster stable and participatory labor-management relations so as to facilitate flexible adaptations in response to changing economic environments. Industrial relations characteristics can hinder the development of stable and participatory labor-management relations systems. Industrbal Relatlons Characteristics Although industrial relations systems in developing couintries are diverse, they often share certain common characteristics that inhibit the development of successful industrial relations practices. An illustrative (not exhaustive) list of such characteristics includes a central role for the state in economic planning and industrial relations regulation, weak and political unions, unwieldy dispute resolution mechanisms, inappropriate union and bargaining structures, the frequent absence of genuine collective bargaining, and poorly educated workforces. The State and Industrial Relations Policy. The state is the major actor in the industrial relations systems of developing countries, and from a conceptual standpoint, state policy can be classified into two primary approaches, the "equity approach" and the "efficiency approach." It must be remembered that there are substantial variations in third world industrial relations practices, and that these approaches are perhaps oversimplifications, but are used here for illustrative purposes. Many countries have elements of both approaches. The "equity" approach accepts the principles of free collective bargaining and is characterized by protectionist labor legislation and social welfare policies. Protective labor legislation is extensive under this regime, and impinges on the employer's cost and flexibility. Typically, protective legislation includes: restriction on layoffs with substantial layoff compensation to be paid (e.g., half salary for first 90 days in India); govemment-required permission to close plants; the requirement to give unions 6 extensive notice before altering plant operations; unfettered trade union rights: wEltfare regulations mandating the provision of food at subsidized rates and day care facilities at all Industrial factories; and legislation mandating extensive employer-paid benefits. The motivation for the equity approach often represents a political bargain between unions and ruling governments, who were perhaps partners in the struggle for independence, but also reflects reaction to populist pressures, and the need for legitimacy gained from ILO membership. A partial justification for the equity approach is that in the absence of govemment-run social security and unemployment compensation programs, and weaknesses within labor movements, employers must bear the cost, and labor should be protected. Adoption of an import substitution industrialization policy, with substantial protection for domestic industries fom intemational competition, further reinforces the equity approach. Many former British colonies in Asia and Africa exemplify this approach. It is important to note that the equity approach is not always equitable within each country, as these policies often apply only to the modem industrial sector, often a very small part of developing economies. The "efficiency" model, on the other hand, provides entrepreneurs with considerable. freedom over labor issues. Although labor welfare policies (working hours, minimum wages, prohibition of child employment) are perhaps similar to the policies of countries following the equity model, in the efficiency approach trade union rights are restricted. Examples of these restrictions include: prevention of unions from political activity; curtailment of the scope of collective bargaining (unions restricted from bargaining on promotions, transfers, job assignments, layoffs, hiring, etc.); prohibition against strikes; banning of unions in certain areas of the economy (e.g., in export processing zones); and noncompliance with various ILO resolutions regarding the freedom of association. The efficiency model appears prevalent in countries that have adopted an export-oriented industrialization policy during the 1970s (e.g., South Korea, Taiwan, Malaysia, Philippines) (Deyo 1989). It is important to note that countries often follow both approaches simultaneously. Many countries enact protectionist and welfare-oriented policies consistent with the equity approach, but restrict trade union rights. African examples of this duality include the following: govemments in Zambia and 7 Burkina Faso have arrested union 'oaders: the Nigerian government has restructured unions and the Nigerian labor congress was dissolved in the 198Ws; wage ceilings have been imposed in Tanzania, Nigeria, and Kenya; the subject of bargaining has been rstricted In Zambia; and many other countries have introduced structural reforms of trade unions via legislation designed to reduce their influence (e.g., Algeria, Egypt, Ethiopia, Nigeria, Tanzania and Zambba). Both approaches have their problems. The equity approach can be praised for its enlightened social protection and adherence to the principles of freedom of association enshrined in ILO conventions. However, it can be criticized for is excessive labor protection, excessive poliicization of unions, unfettered trade union rights, and imposition of social costs on the employer that impinge on competitiveness and adaptability to changing environments, The efficiency approach Is advantageous since it often appears to be integrated with industrialization strategies necessary for rapid econornic development. Yet, i; can be faulted for Uts labor suppression, which, over the long term (e.g., South Korea) results In conflict and the destabilization of industrial relations. What is important, and what third world countries have not achieved, is an industrial relations policy that balances the two approaches and is Integrated with the broader economic and social policies that are needed to meet development objectives. Unions and Politics. The politial character of unions has been well documented (e.g., Essenberg 1981). In muitiparty democracies (e.g., India), political parties depend on trade unions for support, and in single party democracies (many African nations), unions face pressures to align themselves with the party structure. The weakness of third world unions vis-a-vis employers forces unions to pursue methods to gain what they cannot accomplish via traditional economic and bargaining methods. This paper does not argue that trade unions should not pursue political action to achieve their goals. Rather, we wam against excessively political unions, and the subordination of economic objectives and methods to political objectives and methods. Excessively political unions inhibit the 8 development of stable union-management relations. Two unfavorable outcomes of excessive political unionism, using India as an example, are cited below. Unlon leadership in India is a stepping stone to party leadership. Aspiring party leaders also serve as elected leaders of the local trade unions in firms where they are not employed. Consequently, union policies reflect the priorities of the political party, and are not sensitive to the economic situation of the local firm. It is common, for instance, for political parties to call nationwide strikes for political reasons that have little to do with economic issues (Ramaswamy 1983). Indian law allows multiple unions in enterprises. Since each party has its own trade union wing, there are multiple unions in each enterprise. Since each union strives to increase its membership, there is constant competition for membership loyalty. Workers in the plants shift their allegiance from unions that fail to deliver to unions that promise to do so. Given this interunion rivalry, stable labor- management relations do not develop. For stability to develop, a single bargaining agent in each enterprise is required. Yet legislative efforts in tiis direction have repeatedly failed, given that this is not in the interest of competing political parties and unions. Multi-unionism (along with the constitutional nrocedures that mandate annual union elections) thus typically requires Indian management to negotiate with a different set of leaders at erch contract renewal. Given this, along with the political nature of unions, managements have little incentive to invest in developing long term relationships with any single union, and union leaders have little incentive to pursue long term policies that meet firms' needs. Weak Unions. A number of characteristics interact in most third world industrial relations systems to produce weak unions, which hinder the establishment of genuine and stable collective bargaining. These factors include: the relatively small industrial sector that is usually organized (typicaliv, the unorganized informal and agricultural sectors are much larger than the industrial sector); the relatively uneducated labor force which can be manipulated by leaders who have goals different from those of the members and the firm (e.g., the incidence of nsweetheart contracts" is estimated to be 20 percent in India); the existence of multi-unionism which serves to fragment the labor movement nationally and at the workplace level; the absence of legislation promoting single bargaining agents inside firms and an 9 orderly bargaining structure; repressive state policies that grant decertification and deregistration powers to govemment officials; successful employer opposition to union formation; government persecution of labor leaders; the absence of a unified employers association that could inspire stable industrial level bargaining; and union constitutions that mandate union elections on a yearly basis thus preventing the development of long-term plans by union leaders. Needless to say this list only highlights some of the problems leading to weak unions. It is relevant to note that union densities in developing countries (calculated as a percentage of the non agricultural labor force) ranges from 1 percent (e.g., Upper Volta. Haiti) to about 25 percent (e.g., Argentina), with the mean well under 20 percent (Kurian 1985). Weak unions imply unstable industrial relations as managements use their power to restrict unions and are able to push through agreements that leave workers unsatisfied, leading to shop- floor conflict. Consequently, unions turn to the political resolution of issues that cannot be s3lved via bargaining. As a result, unions rely more and more on dispute resolution mechanisms, particularly since their weakness leads them to push every grievance to a disputa for fear of otherwise losing membership. In this environment, unions cannot take a long-term view of the labor-management relationship as all their energies are focused on sh-ort term issues. Meanwhile, employers have little incentive to deal with weak and therefore, unreliable unions. Thus, there is little development of the trust and cooperation necessary for quality production, flexibility, and long term stable labor relations. DisDute Resolution Systems. An effective dispute resolution system encourages the parties to reach bilateral solutions to disputes, thereby enhancing the development of a stable collective bargaining relationship. Most third world systems, however, have the opposite effect: they produce excessive reliance on third party intervention. Many dispute resolution procedures have been inherited from the colonial occupiers, or have been adapted (often selectively) from successful models in the industrialized world: often they have been designed primarily to stop interruptions in the production process (Damachi 1986). Although the range of mechanisms vary, a few central characteristics are common to developing countries. 10 Most dispute resolution procedures Impose iimitations on strikes. Rather than allow unions to use their economic strength, strikes 2re curtailed. For example, workers cannot continue their strike while a dispute is under conciliation, mediation, arbitraton or is being decided by a labor court. Govemment officials are often given the power to Intervene in disputes and suggest solutions, or even mandate s ilutions. Most conciliation, mediation, and arbitration services are carried out by government departments. This subjects the procedure to political interference (Lansing and Kuruvilla 1987j. Influential political leaders (who may be leaders of unions, or have influence over key ministers and secretaries in the labor ministries of these countries) can often exercise influence over conciliators and mediators. The effectiveness of the procedures suffers from delays as well as the lack of adequately trained officials in govemment dispute resolution departments. In most cases, if the dispute is not settled through mediation at the local level, it is then transferred to the regional level, and finally to the state level of mediation. If it is still not resolved, it Is transferred to the Minister of Labor in each state, or referred to compulsory arbitration, the labor court or an industrial tribunal. Resolution of disputes thus can take months or even years. The authority granted to concilistors and mediators to refer disputes to binding arbitration, labor courts, or industrial tribunals introduces further delay into the dispute resolution process. The weakness of unions actually encourages the use of such procedures i.e., a "narcotic effect". Unions tend to rely on dispute resolution procedures in order to obtain what they cannot obtain through direct collective bargaining. In situations where multi-unionism exists, the dependence is much higher, given that unions have to constantly raise issues to attract membership, as a result of union competition. The Pressures on Industrial Relations In response to economic pressures, most developing countries (and this applies equally to the middle income countries of Eastem Europe) face pressure to liberalize their economies. This liberalization implies a change in industrial strategy, from an isolationist import-substitution industrialization 11 (1S) policy with Its attendant protectionist programs (see for example, the current liberalization in the Indian economy), to a more export-oriented economy that is integrated with the world market. This transition Is apparent In Asia where many countries (e.g., Malaysia, Philippines, Indonesia, Thailand, India) have shifted or are shifting into more export-oriented strategies. The transition is less apparent in Africa, where there Is nonetheless evidence of moverrent in this direction. The integration of these export- oriented economies into the world marketplace has substantial implications for industrial relations. Export-oriented development requires industrial relations systems that are flexibleand able to adapt to change. For many developing countries, this implies a drastic restructuring of industrial relations policies at the national, firm, and work place levels. If third world economies are to successfully integrate and compete in the world market (and World Bank policy prescriptions push them in this direction), they have to compete with modem production and industrial relations practices. Th;s implies that late developers cannot just follow the development paths of the advanced countries with a lag, such as the traditional U.S. based mass production (Fordist) strategy. Instead, developing countries must quickly create more flexible systems. For example, many of the NIC's involved in export oriented industrial relations have had to adopt production techniques similar to those used in the industrialized world in order to compete successfully. The requirement to compete with best practice implies a "quantum leapu for most countries, and underscores the urgent need to address industrial relations issues. Although the organized sector is very small compared to the unorganized and informal sectors of most developing economies, the need for flexible industrial relations policies is still paramount, given that it is the organized sector that plays a central role in industrialization, exports, and structural adjustment. This is not an easy task, since there are often compelling reasons for the existence of inefficient industrial relations characteristics in the third world. In equity based systems, for example, laws limiting employer flexibility (e.g., limiting employer freedom to layoff, retrench or close their industries) exist, in part, because of the absence ef altemative sources of support from the state (such as unemployment insurance policies, social security systems, and labor market and training institutions). 12 The lack of resources to establish social security, for example, also explains why employers are forced to provide extensive retirement and health benefits. As a result, the establishment of appropriate industrial relations requires a rethinking of other national policies. A national industrial relations policy is useless if viewed in isolation; it must be linked to broader policies regarding education, training, social security, etc. This broader linkage between industrial relations and other policies has by and large been ignored by policymakers. Below we examine in detail some of the basic requirements of any industrial relations system. We start with basic principles that should be followed in industrial relations and then discuss some of the problems that have appeared in developing countries conceming these industrial relations features. C. Basic Principles of Law and InstitutIons Although there are wide variations in the industrial relations characteristics of advanced industrial economies, successful countries appear to share certain basic principles of law and institutions germane to industrial relations. The Need for Unlons, the Right to Strike, and Collective Bargaining Modem industrial relations requires that workers be afforded the basic right to form unions and those unions should be granted the right to bargain collectively with employers and to strike if a negotiated settlement is not reached. The social advantages gained from the provision of those rights are revealed in the experiences of industrialized countries. The advantages include the fact that workers are entitled to a say in the determination of their employment and working conditions. Not only is such input justified on rights grounds, but in addition, contemporary experience shows that to compete in world markets requires the active participation and commitment of the work force in order to meet quality and product performance standards. There also are clear virtues derived from industrial relations systems that give the parties involved in the production process direct involvement in the determination of work conditions and 13 employment terms. Employers and workers and their representatives know their own problems best and also can be remarkably adept at devising practical solutions to problems or conflicts. Interference by outside parties In the resolution of problems often leads to apparent solutions that prove unworkable or impractical in the long run. Furthermore, third party interference eliminates the constructive learning the parties receive in problem resolution and identification as they work through their own problems. Even if a third party could impose a solution that solves a problem in the short term, this process prevents the parties from developing the capacity to solve their own problems. Thus a clear advantage to collective bargaining is that it avoids a cycle of dependence on third party interference. Experience in the industrialized countries also shows that unions, the right to strike, and collective bargain.ng are more likely to produce outcomes that are acceptable to workers and employers as compared to solutions that are imposed by govemments or other third parties. Not only are employers and workers more likely to develop solutions that meet their own needs, these parties are also more likely to feel a commitment to making such solutions work. Imposed employment terms in contrast are often resented and resisted in part because the parties who must live with these terms had little say in their development. Thus, the process through which employment terms are set is often as important as the actual terms themself. Collective bargaining has the advantage of being a process that is "owned" by the parties involved. Since collective bargaining is a healthy process, it should be applied widely. This suggests that the right to strike should be granted in nearly all cases. There may arise some instances where an altemative to the right to strike is necessary, yet experience demonstrates that the use of strike altematives should be limited. It makes sense, for example, to exempt certain public services, such as police, firefighter, or emergency medical care, from the right to strike. Experience in North America shows that in these cases the availability of interest arbitration (where a third party determines the terms of the labor contract if an impasse is reached in negotiations) can be a successful altemative to the strike (Olson 1988). 14 A country may also benefit by providing a national emergency dispute resolution procedure to limit the right to strike on a case by case basis In industries other than those listed above. The United States has such a procedure provided through the Taft.Hartley amendments to the National Labor Relations Act and these procedures have worked well (Cullen 1968). Yet, such national emergency procedures should be narrowly applied so as to avoid regular Interference into the normal process of collective bargaining for all the reasons outlined above. Care should be taken in the construction of emergency dispute regulations to avoid the excessive use or influence of such procedures. Independent Mediation Entitles A very constructive role can be played by independent mediation entities in a modem industrial relations system. As recommended above, in nearly all cases unions and employers should be allowed to settle contract terms through direct negotiations that allow unions to engage in strikes if negotiations reach impasse. Yet, third party mediation should be available to help the parties in their negotiations. The proper role of mediation is to facilitate communication between labor and management and provide advice. In the United States such mediation comes through the Federal Mediation and Conciliation Service while in Britain the Advisory Conciliation and Arbitration Service meets these needs. Mediation should be accomplished in a manner that preserves the central place of direct collective bargaining between labor and management. Mediation should not acquire the power to impose settlement terms as this merely becomes a vehicle for third party interference in bargaining (Feuille 1979). For mediation to serve such a function there must be a sufficiently well trained cadre of mediators who understand practical labor relations issues. Countries that have recently experienced the spread of democratic political institutions especially need to devote resources toward the training of such mediators. There also is an extremely healthy role to be played by grievance procedures or labor courts that solve problems that arise during the terms of contractual agreements. Research shows that grievance procedures can provide employee voice and thereby lower employee tumover as well as assist 15 in the Identification of problems and informal resolution of conflicts (Freeman and Medoff 1984). Successful grievance procedures are designed by the parties who directly use them and not by govemment or other third parties. In Europe, govemmentadministered laborcourts commonly resolve individual complaints. Labor courts, like grievance arbitration, can effectively settle problems and provide practical resolution of problems and avoid larger social conflicts. It is possible for an industrial relations system to contain both grievance arbitration and labor courts, yet most national industrial relations systems emphasize one or the other. Proper Unlon Robe The earlier discussion highlights the pressing nature of current intemational competitive conditions. There is clearly much for employers and unions to focus on conceming how the work place can produce employment terms that are socially acceptable and products that are competitively successful. This does not imply that there is no role for unions to play as polkical actors or participants in political debate. It does suggest, however, that the representation of worker interests often requires unions to focus their attention on employment and work place issues. It also is clear that socity benefits trom the presence of politically independent unions that are free from domination and govemment control. Such domination eliminates the democratic representation of worker wants and constructive employer-union interchange. The drawbacks from state domination of unions are many including a violation of worker rights. Furthermore, state dominated unions or state controlled industrial relations systems tend to generate explosive periodic conflict. Such has been the case in South Korea and some countries in Latin and South America (Deyo 1989). As representation and bargaining entities, unions should not serve functions that are properly the don r of govemment social service institutions. For example, there are proper social service needs in the areas of social pensions (e.g., social security) and vocational education that should be regulated and financed by national govemments and/or regional govemments. Unions not only are 16 poor providers of such services, they also become diverted away from the areas where they can make more substantial private and social contributions by pursing social welfare provision activities. We do not intend to convey that there is no useful role to be played by unions in the provision of any social services. In some countries (e.g., Sweden and Denmark) unions, for example, play a constructive role in the administration of unemployment benefits. And in many countries unions are effectively involved in vocational education programs. Unions also should not become the vehicle for the direct provision of consumer goods such as housing or food distribution as they did in the former Soviet Union. Not only are unions relatively inefficient as a distribution and pricing system, consumer goods provision through the union or work place also unfairly and inefficiently ties workers to specific firms. These concerns do not mean that unions should be blocked from addressing specific worker health or educational issues. Union administered health and welfare funds have played constructive roles in many countries (e.g., Singapore). The point is that social programs should in general be administered by governments, while private programs should have narrower bounds and focus. Workplace Representation There are great advantages to an industrial relations framework that includes parallel representation at the workplace through institutions such as works councils. In the most successful examples (as in Germany, the Netherlands and Denmark), basic pay and work standards are set in collective bargaining at an industry level (for an entire industry or group of industries, either regionally or nationally). Plant and firm-level works councils, composed of elected blue and white collar representatives, enforce the collective bargaining agreements and also engage in ongoing discussion and negotiation of plant-specific issues. Well functioning works councils typically have formal rights, backed up by national legislation or central collective bargaining, to extensive information regarding company 17 plans, as well as to consultation and in some cases (usually personnel issues) veto rights in management decision making. In Germany, works councils have integrated employee representatives into management decision making and provided a flexible instrument for restructuring and adaptation. Although works councils have imposed some extemal rigidity on firms (i is more difficult to hire and especially to fire), this has been more than compensated for by added intemal flexibility (Streeck 1984a). Because they are able to negotiate relative employment security for the workforce, works councils are more willing to accept internal flexibility of deployment (Katz and Sabel 1985). Works councils also push firms to increase in- house training and the hiring of apprentices. Although management decision making can be slowed in the legally-established, consensus-building process, managers have important allies in the works councils to smooth implementation once decisions are reached. Works councilors at large plants often take on co- management functions, relieving management of important personnel responsibilities (Turner 1991). Works councils also provide a voice for white-collar employees who are traditionally underrepresented in industrial relations systems; and the councils provide some representation for employees even where unions are weak or absent. Where works councils function well, they contribute to a virtuous circle that includes employee voice on important issues, comprehensive representation, labor-management cooperation (since the works council identifies with the interests of the firm), high trust based on joint efforts, flexibility in labor deployment, and high productivity and product quality (Streeck 1984a). These plant-level institutions of codetermination make i possible for works councils, unions (often working through the works councils), and managers to build successful "productivity ct)alitions." it is important to emphasize, however, that the virtuous circle is possible not just in isolated cases (as in the U.S. and Britain) but throughout the economy (as in Germany), and only because works councils are established and defined in national legislation and/or a collective bargaining agreement. 18 The parallel representation afforded by works councils Is also buttressed, and made more successful, by arrangements at the strategic level. The i tter include both collective bargaining between Industrial unions and employer associations and uniontworks, council participation (also mandated by law in Germany) on company supervisory boards. The presence of strong unions and employment security in Germany also help to make works councils successful. Enterprise unions at large firms fulfill some of the works council functions in Japan. These include co-management, facilitating labor-management cooperation, building productivity coalitions, receiving extensive Information regarding company plans, consultation, and giving a voice to white-collar concerns. Enterprise unions are also integrated to some extent into management decision making through firm-level bodies such as joint labor-management committees. But as in the U.S. case, this form of plant-level representation has no legal backing in formal participation rights and is weak or non-existent for the majority of workplaces which are non-union. D. Waoo Settlnr and Baraalnlna Stmueture Two related issues that are of central importance in industrial relations in any country are the structures for wage setting and bargaining (the latter is referred to as bargaining structure). Normally, there is a close connection between these matters as wages are set as part of the normal bargaining that occurs between employers and unions. Bargaining Structure A critical choice in the process of collective bargaining is the degree to which the bargaining structure is centralized. The most extreme case of centralized wage setting (w',ich does not involve collective bargaining) occurs during incomes policies when a govemment imposes pay standards that apply to all workers in the economy. Where there is no govemment sanctioned incomes policy, the common most icentralizedK bargaining structure occurs when a large employer association negotiates with a union (or federation of unions) and sets wages and other employment terms for all the unionized employees in all the firms that are members of this employer association. Bargaining in Sweden until 19 recently was a frequently cited case of centralized bargaining involving the LO and SAF. Mid-range degrees of centralization in bargaining occur where a single labor contract is applied to all the unionized workers in all the plants of a single company or industry. The most decentralized case is where wages and other terms in the labor contract are set in a contract that covers only the workers in a single plant (or a subset of the york force in a plant). The D1advantages of Fragmented Bargaining Structures The challenge in developing countries is to develop wage setting and bargaining structures that meet the parties' needs and the pressures of intemational competitive forces. For instance, highly fragmented bargaining units should be avoided. A number of problems have appeared in industrialized countries in situations where bargaining units were too fragmented and numerous. In Britain, for example, the presence of multiple craft and industrial union bargaining is said to be an impediment to integrstive problem solving and the effective introduction of technological change (Ulman 1968). A large number of bargaining units also contributes to inflationary pressures through the promotion of wage imitation and leapfrogging across jurisdictions (Flanagan, Ulman and Soskice 1983). The Virtues of Coordinated Bargaining One potential solution to the bargaining unit fragmentation problem is to conduct wage bargaining at the level of industrial sectors. In Germany, for example, bargaining occurs within 16 major industrial sectors. The German structure has the advantage of inducing unions to consider broad worker economic interests. So for example, because unions in Germany often include workers from a variety of companies and industries, there has been less trade protectionist sentiment within German unions as these unions often include members whose livelihood depends on the export success of their employers. In this way, particular firm or industry interests become muted in the bargaining process (Streeck 1984a). The coordination across work groups provided in the Gerrnan bargaining structure meshes with other 20 aspects of German Industrial relations to produce was has been called a "coordinated market economy" (Soskice 1990). Another successful altemative bargaining structure Is enterprise bargaining of the sort that occurs in Japan. The advantage of enterprise bargaining is that it is very responsive to the economic conditions facing particular firms. The potential problem of inflationary leapfrogging across enterprise unions is dampened in the Japanese system through the simultaneous occurrence of wage bargaining that occurs in the "Spring Offensive" each year. In the annual offensive the various enterprise unions communicate extensively with one another and often discuss wage demands with their respective union federations. In this process, wage demands are coordinated and take account of macroeconomic circumstances. Japanese enterprise wage bargaining also has the advantage of often including a sizeable share of pay increases in the form of annual bonuses that are influenced by the economic per!ormance of the firm. The advantage of linking pay directly to economic performance through this and other altemative techniques is discussed more extensively below. The Virtues of Annual Pay Agreements Both the German and Japanese wage bargaining have revealed relatively little inflationary tendencies. This appears to follow, in part, from the fact that in both countries most wage bargaining occurs on an annual basis, namely, the wage terms of labor contracts extend for only one year. In addition, in these countries there are various processes which provide coordination across wage bargains. These coordinating mechanisms seem to be especially critical in avoiding inflationary pressures (Soskice 1990). In contrast, multi-year labor contracts in the United States (and some other countries) reduce the influence of contemporary macroeconomic factors on wage negotiations. In this way, multi- year contracts contribute to wage rigidity, a particularly worrisome problem during economic downtums (Wachter 1976). A countervailing advantage to multi-year contracts comes from the fact that they limit the possibility of costly strikes that might ensue in more frequent bargaining, in effect, by reducing the 21 costs and risks of bargaining. Yet, the inflationary tendencies inherent In multi-year contracts appear to outweigh this virtue. The Occupational Scope of Bargaining Structure Another important aspect of bargaining structure concerns which employees are included in unions and other representation structures. In Japanese enterprise bargaining, white collar employees of the firm, below the executive rank, are included as members of the union. In Germany, whie collar employees often press their interest through their participation in the codetermination processes which provide these employees with the right to proportional representation on the works councils and in the election of supervisory board members. In addition, white-collar employees in Germany belong to the same industrial unions as blue-collar workers. The provision of representation rights to white collar employees is an aspect of bargaining structure worthy of attention. Both the German and Japanese methods of providing white collar employees with representation rights are commendable. An important industrial relations task confronting developing countries is to avoid representation structures that only include "blue collarn employees. Recent Movement to More Decentralized Bargaining Structures In recent years in very many industrialized countries there has been movement tcward more decentralized formal pay setting and bargaining structures. In addition, in nearly all countries there has been a shift toward bargaining structure decentralization through the elevation in the importance of, and an increase in the variation of, the outcomes of plant level bargaining. Greater variation appears particularly in the work rules and work practices agreed to by union and company officials at the plant level. Labor and management have turned to more decentralized bargaining as a way to develop contract terms that respond to their needs and economic pressures. As well, more decentralized 22 bargaining fits well with the shift towards new industrial relations practices such as team systems discussed later In this paper. The current worldwide movement towards decentralized bargaining structures are evident in a few developing countries as well. During the 1980s, Singapore, Taiwan, Malaysia, Philippines and South Korea have adopted elements of the enterprise union system. While Singapore has achieved, like the Japanese, a decentralized system buttressed by some degree of coordination at the macro-level through national wage councils, South Korea is moving towards a similar arrangement. Mal, ysia permits enterprise unions, within an industrial union framework that allows enterprise unions to respond to their firms unique economic circumstance, but are able to rely on their industrial affiliates for support. These developments, however, are the exception rather than the rule in the third world. While it is perhaps early to evaluate the efficacy of these systems, unions still remain waak and fragmented in South Korea, Malaysia, and the Philippines. Furthermore, it is unclear whether the benefits of decentralization observed in Germany, Sweden, and other nations will be realized without attendant improvements in union security and related policies in developing countries. The direction of structural change in Africa reflects trends in the opposite direction. Restructuring in many African countries (such Algoria, Ethiopia, Egypt, and Nigeria) has produced more centralized monolithic union structures, with the attendant problems. Yet, recent (1991-1992) political instability connected with movements towards more genuine forms of democracy may reverse this trend. By and large, the experience of the industrial countries suggests that industrial relations outcomes that reflect economic circumstance require decentralized bargaining that is over the long run supported by strong unions and stable labor-management relations. E. WorkDlace Issues There have been numerous recent changes in work place industrial relations in the advanced economies. 23 Recent Restructuring In Industrial Relatiorn In some countries recent changes amount to a fundamental transfotmation in industrial relations. It virtually all countries, aspects of this transformation have appeared. There are many dimensions to this transformation including a shift in the focus of industrial relations activity away from the collective bargaining to the *strategic' and *workplaceu levels (Kochan, Katz and McKersie 1986). In contrast to the traditional arms-length and formal nature of collective bargaining, the "new industrial relations system involves more continuous and informal relations between workers and managers (Windolf 1989). The new system often Includes contingent compensation, team systems of work organization, enhanced worker and union participation in decision making, and employment security. Contingent Compensation Contingent compensation links pay directly to firm or worker performance. The virtue of more contingent compensation comes from the fact that it provides responsiveness in wages to current economic circumstances. As Weitzman argues, if pay quickly responds to economic conditions, economic systems are able to lessen the need for large quantity adjustments in employment and output and thereby become less prone to stagflation and related macro economic problems (Weitzman 1984). As discussed earlier, the presence of annual pay agreements rather than multi-year pay agreements adds a significant degree of responsiveness. Yet it appears that economies benefit further from the presence of direct linkages between pay and measures of firm and/or worker performance. The use of annual bonuses that provide one-third of total wage eamings in Japan provides a virtuous high degree of pay responsiveness to economic conditions. It should be noted that for these bonuses to provide a macro economic stabilization function they must vary with firm performance which does occur in Japan (Hashimoto 1979). Other systems would be well served to develop mechanisms that fit within their own fr^,.ework to provide pay responsiveness. 24 Team Systems A second important Innovation of many new work systems is the shift to various forms of shopfloor and office teamwork (Windolf 1989). Traditional assembly-line organization based on isolated work stations, it tums out, contains built-in limitations on worker participation and flexibility. Traditional 'fragmentedt job assignments limit worker contributions to production improvements, commitment to product quality, and capacity for flexible deployment in the workplace. Japanese team organization, by contrast, including cross-training, job rotation, integration of tasks, and regular team meetings, has demonstrated dramatic potential for productivity and product quality gains. At the same time, unions in Germany, especially IG Metall, have been promoting "group work," both for its production improvements and as a more humanistic form of organization (Tumer 1991, pp.111-17). The German version places an emphasis on broad work assignments, broadened responsibility and autonomy for the groups, as well as enhanced skills training and broad upgrading of group members. Although experience with teamwork in the U.S. and Britain shows mixed results, evidence from Japanese plants, from Japanese-run plants in the U.S. such as NUMMI, and from German pilot projects indicates that considerable gains are possible in productivity, quality, and flexibility, with or without newtechnology (MacDuffie and Krafcik 1992; Katz, Kochan and Keefe 1987). Gains are especially likely when new work organization is associated with extensive skills training, so that workers are able to integrato tasks and solve problems that arise, and with cooperative, participatory labor-management relations. Team or group systems are thus most likely to work well in countries such as Japan and Gerrnany, that already have strong institutional underpinnings for training and for worker/union participation. In other countries, it is necessary to establish such underpinnings, to promote cooperation. participation and training, along with the introduction of more flexibleteam or group forms of organization. The necessary underpinnings can range from joint labor-management committees at the plant level to national legislation to promote training and to establish some nationally appropriate version of works councils. 25 Employee Involvement In Decision Making A third essential element of innovative work systems and "new industrial relationsu is mechanisms for employee and union participation. The relative success of Japanese and German industry has brought home the virtues of employee and union participation at various levels of management decision making. As changing world markets and new information technologies have increased the need for functional flexibility in the workplace, production requirements can no longer be met only by armies of semi-skilled workers following orders. Japanese and German industrial relations systems, in different ways, have offered strong mechanisms whereby (i) workers, unions and/or works councils actively contribute their own ideas to improving the production process; (ii) employees have a better understanding of the broader production process and are more suited for flexible d .ployment; and (iii) workers, unions and/or works councils receive advance information regarding company plans and participate in decision-making processes, and thereby often support company plans and help smooth implementation. In an era in which innovation in both product and process has become increasingly important, institutionalized participation often means that employees and their representatives become allies of management (rather than enemies or obstacles) in necessary work reorganization. A further push for increased participation comes from the spread of team systems which provide wider roles for hourly workers as they become involved in production control tasks. Team systems and the administration of new employment security are also often associated with a broadening of worker and union involvement in strategic business decisions. In part this strategic involvement arises as a consequence of the major corporate reorganizations that confront the workforce. After watching plants close and work outsourced, workers and unions have sought avenues to affect the decisions that weigh so heavily on their future. Difficulties In DIffusing Participation The U.S., Britain and France, among other countries, have weak participatory traditions, with industrial relations systems that do little to encourage either participation or intermal flexibility. 26 Historically, managers act (no one tells me how to run my business"), unions react (and often fight rearguard actions against innovation), and workers are supposed to do what they are told (but often sabotage innovation in more or less subtle ways). In these countries, a major challenge is to reform industrial relations practices so that new participation and flexibility are possible, in line with the demands of contemporary world markets. In the face of entrenched insttutions and practices, however, such reform is quite difficult. Reform is underway In all three countries, but the pace is halting and the failures are many (Milkman 1991). Japanese and German firms, by contrast, have been lucky: firms in both countries operate within institutional contexts that have tumed out to offer participation and flexibility at an historical juncture where such attributes contribute to market success (Katz and Sabel 1985; Soskice 1990). As American, British and French managers have noted, various kinds of employee participation and functional flexibility have been shown to improve productivity, lower costs, improve product quality, and increase the capacity of firms to change product offerings quickly. Cases of successful reform and innovation within each of these countries has further brought home the lesson (Eaton and Voos 1992). From quality circles to shopfloor teamwork to joint labor-management committees to financial participation to increased employment and income security, innovations have been spreading in unionized and non-union workplaces alike. The crucial dilemma facing firms, unions and policy-makers outside Japan and Germany is not how to import foreign models but how to adapt and reform established institutions and practices to achieve the necessary participation and flexibility. Participation can be structured in a number of different ways, including: works councils established by law, informally established cooperation through enterprise unions; joint labor-management committees established through collective bargaining; and quality circles and suggestion programs in union and non-union workplaces. It is important to note that where unions are present, they need to be fully aboard and supportive of the introduction of new work organization. Management strategies that divide the local 27 union to force teamwork on a reluctant workforce are unlikely to work (Turner 1991). Institutional arrangements that ensure union participation, at least at some level, can facilitate the labor-management cooperation necessary for the high-trust, innovative modern workplace. While successful in Germany and Japan, the patchy track record in the U.S., Britain and France suggests that participation requires either legal backing or substantial protections and encouragement throughout the firm. Workers and unions need institutions with clear rules and protections to allowthem to participate in firm decision-making processes: management needs thorough- going reform from the top of the or-anization to the bottom to instill commitment for employee participation. Both sides need training in participatory processes and trust building to set successful participation in motion. The role of middle management is especially important in this regard. It is here that employee participation is most often perceived as threatening to established prerogatives, and it is at this level that participation often fails or is sabotaged. Yet middle managers have a crucial role to play in participatory processes and can find their influence within the organization enhanced rather than diminished. They therefore need extra training in the management of participation, along with special encouragement, protections and clarity about their own (often quite new) roles. Participation is most likely to succeed if it occurs and is reinforced at all three levels of labor-management interaction in the firm: at the strategic level, in r'gotiation and agreement between top management and union leadership; at the functional level, in collective bargaining agreements and contract enforcement; and at the workplace level, in team or group organization and other meaningful participation programs that have the full support of the organization and its management. National labor policy can play an especially important role in encouraging the spread of participation, through efforts ranging from education to specific incentives to enabling or even mandating legislation. 28 Employment Security There appears to be a direct link between participation, employment security and workforce flexibility within the firm. To the extent that workers participate in decision-making processes, they become more flexible in what they are both able and willing to do in production processes. And to the extent that they have employment security, employees are often willing to go along with even the most radical shopfloor innovations. The generation of employment commitment to company goals requires that employees face the prospect of continued employment with their firms. This is particularly important in light of the spread of team systems and the development of more firm specific skills. In the presence of these skills, employees become less attractive to the extemal labor market (Osterman 1988; Marsden 1990). If employees do not then receive more security from the firm, they may well be reluctant to acquire more extensive skills and training, or participate in the sort of shop floor decision making that is so crucial to the maintenance of competitive and high quality products. Employment security can come in a variety of forms. In Germany, there has been a longstanding and successful requirement that if firms seek to reduce the size of their workforce they must first negotiate social plans with their work council. These social plans then outline the compensation received by redundant workers and establish guidelines conceming which workers are to be laid off and create other adjustment strategies. As a result, the German system also has shown relatively less use of layoffs as an adjustment strategy in the face of output declines. The "intemal flexibility' provided in the German system through broad jobs and flexible work rules has created the possibility of more intemal adjustment rather than recourse to extensive extemal adjustments through layoffs (Sengenburger and K6hIer 1987). In Japan, there is heavy use of the "lifetime employment principle". A relatively high percentage of employees spend long careers, often their whole career, with one firm (Cole 1979). The key is that employees are broadly trained and work rules are relatively flexible. This makes it possible for firms to make intemal adjustments to respond to reductions in product demand. These intemal 29 adjustments Include moving workers across parts of the enterprise, or in some cases even "loaning' workers to other firms. The use of intemal movements is, of course, not without its limits. In Japan, lifetime employment is afforded to 'permanent' employees but is not provided to utemporary' employees. In the German system as well, the use of foreign workers (particularly In the 1960s and 1970s) on limited term employment contracts also helped facilitate the less frequent layoff of native German workers (Streeck 1984a). Other forms of employment security, ranging from explicit employment guarantees to retraining and income support exist in a number of countries (e.g., Italy, Belgium, Netherlands), and in many cases, the negotiation of such policies is linked to introducLion of more flexible work rules. The argument here is that labor and management should try to find mechanisms that fit with their respective institutions to make use of intemal flexibility as an altemative to external employment adjustments. At the same time, it is unreasonable to believe that all extemal employment adjustments can or should be avoided. The elimination of the possibility of layoffs can (and in some countries has) produced enormous inefficiencies and rigidities. Govemments should also play a role in providing inter-firm and career mobility, particulariy during periods of slack economic growth and in regions or industries that confront abrupt employment declines. In Sweden, for example, there is a long history of successful 'active labor market policies' that have provided such assistance and mobility to workers (Meidner and Anderson 1973). The New Industrial Relations and the Third World There is little evidence of the emergence of the 'new' industrial relations system discussed above in the third world. However, we can see changes in this direction in the newly industrializing countries of Asia. For example, Singapore has gradually moved to a system of considerable worker involvement both in national as well as firm level economic decisions. Enterprise unions have been adopted in many Asian countries as discussed earlier. Furthermore, broad based training practices are 30 slowly taking root in Singapore, Malaysia, Thailand and South Korea, although much of the diffusion of these practices is due to the predom-nance of Japanese multinationals operating in these countries. Legislative frameworks aupporting worker participation in management decisions through works councils and joint labor-management committees do exist in many developing countries, (e.g., India, Anglophone Africa). And in Francophone Africa, unions have recently been provided with seats in legislative rule making bodies in parliament (e.g., Benin, Senegal). Nevertheless, genuine participation has not taken root given the absence of supporting policies and structures. Despite the legal encouragement for employee participation, and the willingness of unions to participate, the problems of weak and fragmented unions, employer resistance, uneven implementation of existing labor legislation, poorly educated workforces, and most importantly, the absence of stable and cooperative labor- management relations, inhibit the development of genuine participation. F. Tralning and Work Force Skill and Career Development New workplace systems in the 1990s and beyond, based on attributes discussed above such as shopfloor and office teamwork, employee participation, and career development in intemal labor markets, cannot succeed without broadly skilled workers capable of continual leaming (MacDuffie and Kochan, 1991). The Need for Greater Investment In Training The economic environment has put a premium on industry's ability to produce goods using batch rather than traditional mass production techniques, and to make changes in product-mix more rapidly. This has generated new demands on industry's training procedures. More flexible manufacturing processes often entail the introduction of sophisticated electronically directed machinery (such as programmable machine tools) and automated storage and retrieval systems. These systems can be most effectively operated by a workforce well informed of fairly advanced mathematics and statistics. The use of team systems fits into this network of new production techniques as the 31 organizational vehicle through which workes take on the responsibilities to monitor and often diret quality control and production processes. The training needs generated by now production and control systems go beyond traditonal requirements. Not only must the worKforce have quantitative analytic skills, it must be skilled in the use of computer and other electronic technologies. Furthermore, team systems require communication and group skills to facilitate the operation and coordination of team tasks. Factory studies show increasing percentages of skilled workers, many of them newly cross-trained (in additional skills), accompanied by declining percentagas of the traditional semi-skilled (see, for example, Milkman and Pullman 1991). Even the latter require new training in organizational skills (for teamwork, participation, job rotation) and/or more substantive skills development for the monitoring of expensive microelectronic- run machinery (Kem and Schumann 1984). Training and retraining programs have therefore taken on added importance throughout the Industrialized world. This is true both because of the rapid pace of technological change itself, which means that employees need frequent retraining, and because of the nature of new work organization and technologies. The problem for firms and govemments is how to build up the necessary pool of skilled workers, how to build up the abilities and flexibility of the less skilled, and how to develop reliable structures and incentives for continual training and retraining. Union involvement in the promotion and operaton of training programs can be a valuable asset in successful upskilling programs. At the firm level, such involvement can encourage cooperative labor-management relations and form part and parcel of the building of a *productivity coalition" (Streeck 1984b) that includes management, union and workforce. Local unions and/or works councils are often well placed to help run training programs in a way that appears fair and wins the trust of the worktorce. The German Training System The German case also illustrates the constructive role that unions can play in the running of a successful national system of vocational education - as part of a productiviy coalition at the macro 32 level. EstAblished and supported by comprehensive national legislation, and funded both by government and employers, Germany's vocational education system offers three-year apprenticeships across a broad range of occupations: blue and whie collar, manufacturing and service, from bakers to machinists to telecommunications technicians to hair stylists to pre-school teachers to equipment programmers. At the national, regional and local levels, the vocational education system is tripartite: employers, unions and govemment all play an active role in establishing and updating occupational groupings and ensuring that training programs keep up with employer and workforce needs. Because unions are broadly integrated into vocational education, they are committed to promoting a highly skilled workforce. This commitment extends from the national and regional levels, where unions help set recruitment targets based on projected skill needs, to the plant and firm levels, where works councils press management to hire more apprentices. The British/American image of craft unions that restrict entry to the skilled trades (to maintain status and pay ievels for the already skilled) is contradicted by the German picture of unions that actively promote widespread skills training. In part this is due to the industrial union structure of representation; equally important Is the top-to-bottom integration of unions Into the running of the vocational education system. Employers have strong incentives to participate in apprenticeship hiring. The in-school training that apprentices receive is govemment funded; and apprentice wages, paid by employers, are low. The employer thus provides on-the-job training and in return gets inexpensive and often energetic and flexible young labor. In addiion, the employer gets a multi-year period in which to work with and observe the young employee as a potential permanent, skilled hire. For employers as a group (and firms participate In the tripartite goveming bodies through employer associations), the national system of apprentice training and vocational education assures a steady flow of skills on the labor market. The nationwide, tripartite, comprehensive system of vocational education has often been cied as a major factor accounting for Germany's industrial success. A highly skilled labor force is clearly a central component of the virtuous circle that has pushed German firms upmarket Into "diversified quality producUon' (Sorge and Streeck 1988). 33 The Extensive Japanese On-the-job Tralning In contrast to the German system, the Japanese rely relatively little on formal apprentices and yet accomplish extensive training inside firms. The Japanese production system includes jobs that are relatively broadly defined and workers are often rotated across jobs throughout their careers. These production techniques are facilitated by the extensive amounts of on-the-job training received by workers. In addition, skill acquisition and work place flexibility are assisted by the high levels and quality of education provided In the Japanese primary and secondary schools. The provision of extensive on-the-job training in Japan is linked to other features of the Japanese industrial relations system. The long tenure of workers in firms provides strong incentives to both workers and firms for investments in training by guaranteeing retums to training investments. In addition, the heavy reliance on seniority-based pay makes job rotation easy and thereby facilitates training across tasks (Cole 1979). First line supervisors in Japan commonly carefully plan workrs' training and career progression. Some of this planning and heavy first line supervisor involvement takes place as part of the performance and associated merit pay that covers blue collar workers in Japanese firms (Koike 1988). Thus, extensive investments in training do not occur accidently, but rather are linked closely to other practices common to Japanese firms (particularly, the large firms). In this way, the Japanese system illustrates the links that exist across various parts of an industrial relations system. Developing countries must evolve their own methods of employee education, both through the education system and through training in industry. G. The Mlsmatch Betwen Social, Educational and Industrial Relatlons Pollees In DeveloDing Countries The previous sections have shown that successful countries have managed to achieve an integration of industrial relations policies with social and educational policies in ways that enhance workforce skill development and facilitate cooperative labor management relations. 34 Developing countries unfortunately, exhlbits littleintegration across policies. Fundamental mismatches exist. For example, lacking appropriate education and labor market policies, Malaysia has a labor surplus at the professional level and a labor shortage at the skilled, semiskilled, and unskilled levels. The Philippines has a labor surplus at the unskilled level and a shortage at all levels of skilled personnel, primarily due to emigration. This skill shortage ensures the continuance of low cost labor intensive production systems. Weak unions, and the absence of policies and structures that promote cooperative industrial relations further ensure the continuance of low cost, low skill production. The absence of unemployment Insurance, social security, and skill development via the education system forces employers in many third world countries to take on costs which in other nations are handled by the government, thereby inhibiting the competitive position of employers wishing to compete in the world mark.t. Aithough many countries have apprenticeship programs and Industrial training systems modeled after Britain and Germany, the quality of training is below standard for various reasons, most notably because of a lack of resources. Developing nations need to restructure education, labor market, and Industrial relations policies to compete effectively. There are movements towards the ntegration of educational and labor market policies in certain countries. Most of the NIC's (Korea, Singapore, Malaysia), for example, are making strides in both upskilling their workforces and ensuring that the education systems meet their industrial strategy needs. South Korea, having accomplished this link, is now working on other necessary social policies such as the establishment of a social security system, although Korea has not yet articulated an appropriate industril relations policy. Singapore has achieved effective links between Industrialization, social, education and industrial relations policies and provides an Interesting lesson. Having invested substantially in education, Slngapore was faced with continuing low labor cost labor intensive investment by foreign investors. In order to attract more high technology based Investment that would result In upskilling, Singapore used its wage policy. Wages were increased by 12 percent for three years, successfully driving out the low cost producers. 35 These are, unfortunately, isolated examples In the third world. Some of these innovations are gradually taking root in more advanced developing countrles, but even there, not in a systematic way. Although resources are a critical constraint, developing countries must make efforts to integrate various economic, social, educational, labor market, and industrial relations policies in ways that will enhance the abiliy to compete in the intemational economy. H. Concluslons The economic pressures facing the third world suggest the need for the restructuring of their economies and industrial relations systems. But change in industrial relations systems cannot be accomplished overnight. History has shown that there are certain key moments of transition in industrial relations systems, after which they get set and are hard to modify. Often these key moments are as a result of legislative changes (e.g., the establishment of the National Labor Relations Act and the emergence of public sector unions after the burgeoning of public sector legislation in the United States). Other factors are also important in facilitating industrial relations system change, such as important historical or economic junctures (e.g., the post war reconstruction in Germany and Japan were significant moments of Industrial relations transition). In the developing world, previous key moments of national transition have been the independence movements in the late 1940s (India, Pakistan), 1950s (Malaysia, Singapore, and many African countries) and the early 1960s. Transitions in industrlal relations have also occurred in the NIC's due to shifts of industrial strategy: for example, the shift to an export oriented industrialization strategy from an import substitution strategy has resulted in far reaching and fundamental changes in industrial relations In Malaysia, Philippines, South Korea, and Singapore. The recent pressures for structural change in the developing world present yet another opportunity for major transitions in industrial relatlons. A key question is whether countries will take advantage of this opportunity. 36 Another important issue Is the need to recognize that various industrial relations policies and procedures fit together and reinforce one another (i.e., the systemic nature of Industrial relations). For example, changes in pay systems have implications for work organization. Work organization meanwhile has strong Implications for employee participation and skill development. The choice then is not just to emulate elements of Industrial relations policy from the experience of the Industrialized countries and make incremental adjustments, but rather to strive for the reconstruction of entire systems so various new policles fit together in ways that reflect the unique institutional background, politics, and history of each country. Therefore, we are not suggesting that Germany and Japan are the best models for the third worid and Eastern Europe. However, we use German and Japanese examples to highlight successful linkages between industrial relations policies. This paper emphasizes the goals of an industrial relations system and the systemic nature of Industrial relations. Most Importantly we emphasize the necessity of coordinating and integrating industrial relations policy with other economic, social, educational, and legal policies. Developing countries must develop their own unique institutional mechanisms to achieve these goals. Perhaps the points made in this paper will serve as a guide to the principles. While It is clear that many often far reaching changes are necessary to bring about the alignment of industrial relations systems with current economic pressures, the process through which change occurs also is important. For one thing, the process of change has to be managed so as provide for the active participation of the parties. Participation can help to achieve the necessary popular acceptance of changes. Experience has consistently shown that the top down introduction of Industrial relations change (whether it Is policy at the national level or systems at the company level) fails without popular acceptance of the need for change and parUcipation in the change effort. For many developing countries, the changes required in industrial relations involve a "quantum leap", implying a long gestation period (our views are based on the establishment of stable and flexible labor management relations in the long term). Therefore, change efforts should have clear long run goals but must also address issues of transiton in the short term. A stage-wise approach to systems of ^best practice3 may be appropriate. 37 Ref.rsnoe Cole, Robert E. 1979. Work. Mobility and Particiatlon, Berkeley and Los Angel"s: University of Califomia Press. Cullen, Donald E. 1968. National Emeraencv Strikes. Ithaca, NY: New York State School of Industrial and Labor Relations, Comell University. Damachi, Ukandi G. 1986. Contemporary problems in Nigerian Industrial Relations. Edited by Ukandi Damachi, and Tayo Fashoyin. Development Press: Lagos. Deyo, Frederic C. 1989. Beneath the Miracle: Labor Subordination in the New Asian Industrialism, Berkeley: University of Califomia Press. Eaton, Adrienne E. and Paula .B. Voos. 1992. 'Unions and Contemporaly Innovations in Work Organizations, Compensation, and Employee Participation, in L Mishel and P. Voos eds., Unions and Ecorusmic Competitiveness, Washington, D.C.: Economic Policy Institute: pp. 173-216. Essenberg, Bert. 1981. The Interaction of Industrial Relations and the PolIical Process in developing Countries: A Review of the Literature. Labour and Societv. Vol 6, No. 1 (Jan-March 1981). Feuille, Peter. 1979. "Selected Benefits and Costs of Compulsory Arbitration', ILRf 33 (October): pp. 64-76. Flanagan, Robert J., David W. Soskice and Uoyd Ulman. 1983. Unionism Economic Stabilization, and Income Policies: European Experience, VWashington, DC: The Brookings Institution. Freeman, Richard B. and James L Medoff. 1984. What Do Unions Do?, New York: Basic Books: pp. 94-110. Hashimoto, Masanori. 1979. "Bonus Payments, On-the-Job Training, and Ufetime Employment in Japan,' Joumal of Politcal Economv, 87 (October): pp. 1086-1104. Katz, Harry. C. 1992. 'The DecentralIzation of Collectwe Bargaining: A Comparative Review and Analysis", unpublished manuscript, NYSSILR, Comell University, July. Katz, Harry C. and Charles F. Sabel. 1985. 'Industrial Relations and Industrial Adjustment in the Car Industry", Industrial Relations 24 (Fall) pp. 295-315. Katz, Harry C., Thomas A. Kochan and Jeffrey H. Keefe. 1987. 1Industrial Relations and Productivity in the U.S. Automobile Industry", Brookinas Paoers on Economic Actitv. Vol. 3, Washington, DC: The Brookings Institution: pp. 685-728. Kem, Horst and Michael Shumann. 1984. Das Ende der Arbeitsteilung? Munich: C.H. Beck. Kochan, Thomas A., Harry C. Katz and Robert B. McKersis. 1986. The Transformation of American Industrial Relations, New York: Basic Books. Koike, Kazno. 1988. Understanding Industrial Relations in Modem Japan, London: MacMillon Press. 38 Kurian, George Thomas. 1985. The New Book of World Rankings. New York: Facts on File Publicatlons. Lansing Paul, and Kuruvilla, Sarosh. 1987. Industrial Dispute Resolution in India In Theory and Practice. Lovola of Los Angeles Intemational and Comoarative Law Joumal. Vol 9, No. 2, pp.345-377. MacDuffie, John Paul and Thomas A. Kochan. 1991. "Does the U.S. Undclnvest In Human Resources? Determinants of Training in the World Auto Industr, Wharton School, University of Pennsylvania, unpublished manuscript. MacDuffie, John Paul and John F. Krafcik. 1992. 'Integrating Technology and Human Resources for High-Performance Manufacturing: Evidence from the Intemational Auto Industry, In T Kochan and M. Useem, eds. Transformina Organizat;ons, New York and Oxford: Oxford University Press: pp. 209-225. Marsden, David. 1990. Institutions and Labour Mobility: Occupational and Intemal Labour Markets in Britain, France, Italy and West Germany", In R. Brunetta and C. Dell'Aringa eds., Labour Relations and Economic Performance, London: MacMillian Press. Meidner, Rudolf and Rolf Anderson. 1973. "The Overall Impact of an Active Labor Market Policy in Sweden', In Uoyd Ulman. ed. Manoower Proarams in the Policy Mix, Baltimore: John Hopkins University Press: pp. 117-58. Milkman, Ruth. 1991. "Labor and Management in Uncertain Times: Renegotiating the Social Contract" in Alan Wolfe ed., The Recenterina of America: American Society in Transition. Berkeley: University of Califomia Press. Milkman, Ruth and Cydney Pullman. :991. "Technological Change in an Auto Assembly Planr, Work and Occupations, Vol. 18 (May): pp. 123-147. Olson, Craig A. 1988. "Dispute Resolution in the Public Sector', in B. Aaron et al., eds., Public-Sector Bargaininq, Washington, D.C.: Industrial Relations Research Association: pp. 160-188. Osterman, Paul. 1988. EmDlovment Futures: Reorganization, Dislocation, and Public Policv, New York and Oxford: Oxford University Press. Parker, Mike and Jane Slaughter. 1988. Choosing Sides: Unions and the Team Conceot, Boston: South End Press. Ramaswamy E. A. 1983. Indian Management Dilemma: Economic Versus Political Unionism. Asian Sumey (Berkeley) Vol 23, No. 8, August, 1983 pp. 976-990. Sachs, Jeffrey 0. 1989. Developing Country Debt and Economic Performance. Volumes 1 and 2. University of Chicago Press: Chicago, London. Sengenberger, Wemer and Christoph Kohler. 1987. "Policies of Workforce Structure in the American and German Automobile Industry*, in Roger Tarling, ed., Flexibility in Labour Markets. London: Academic Press. Singh, Aji. 1992; The Lost Decade: The Economic Crisis of the Third World in the 1980s: How the North Caused the South's Crisis. Contention:Debates in Society. Culture, and Science, Vol 103, (Spring), pp. 136-168. 39 Sorge, Amdt and Wolfgang Streeck. 1988. Industrial Reations and Technical Change: The Case for an Extended Perspectlve. In R. Hyman and W. Streck eds., New Technoloov and Industrhl RebtLnsa Oxford: Basl Blackwell: pp. 19.47. Soskice, David. 1990. ^Ronterpreting Corpora_s and Explaining Unemployment: Coordinated and Non-coordinated Market Economies', In R. Brunedt and C. Del'Aringa, eds., Labour Relations and Economic Perlormnce London: MacMUlian Pre. Strck, Wolfgang. 1984a. Indusftl Reltions In Ws Gmanv: A Case Study of the Car Indusi, New York: St. Martin's Pre. Streeck, Wolfgang. 1984b. NeoCorporatiut Industrial Reltions and the Economic Crisis In West Germany,' In John H. Goldthorpe ed., Order and Cnflict In Contemoranrv Ceoltalism, Oxford: Oxford University Press: pp. 291414. Tumrn, Lowell. 1991. Democracy at Workg: hanaina World Market and the Future of Unions, Ithaca, Now York: Cornell Universty Pres. Ulman, Uoyd. 1968. Collective Bargaining and Industrial EfficIencyy in Ronald E. Caves and Associates, Britain's Economic Prosoeitv, Washington, DC: The Brookings Institution. Wachter, Mihhael. 1976. wTh Changing Cyclical Revoansiveness of Wage InationO. Brookinas Papers on Iconomig ActMt. Vol. 1, Washington, DC: The Brookings Insitution: pp. 115-59. Wekzman, Martin L 1984. The Shr Econom Cambridge: Harvard Univesty Press. Windolf, Paul. 1989. ProducvMty Coalitins and the Future of Corporatism*, Industrial Relations, Vol. 28, (Winter): pp. 1-20. Policy Research Working Paper Series Contact Title Author Date for paper WPS1 082 What Do Governments Buy? The Shantayanan Devarajan February 1993 C. Jones Composition of Public Spending and Vinaya Swaroop 37699 Economic Performance Heng-fu Zou WPS1 083 Finance and Growth: Schumpeter Robert G. King February 1993 D. Evans Might Be Right Ross Levine 38526 WPS1 084 Stock Market Development and Dong He February 1993 P. Infante Financial Deepening in Developing Robert Pardy 37665 Countries: Some Correlation Pattems WPS1085 Economic Approaches to Modeling Cristino R. Arroyo liI February 1993 0. Nadora Fertility Determinants: A Selective Review 31091 WPS1086 Teachers' Salaries in Latin America: George Psacharopoulos February 1993 G. Psacharopoulos A Comparative Analysis Jorge Valenzuela 39243 Mary Arends WPS1087 Exchange-Rate-Based Stabilization: Alberto F. Ades February 1993 R. Luz Tales from Europe and Latin America Miguel A. Kiguel 34303 Nissan Liviatan WPS1 088 A Primer on the MFA Maze Riccardo Faini February 1993 D. Ballantyne Jaime de Meio 37947 Wendy Takacs WPS1089 Equity Portfolio Investment in Stijn Claessens February 1993 R. Vo Developing Countries: A Literature Survey 31047 WPS1090 Government Expenditures as a Thanos Catsambas February 1993 A. Correa Citizen's Evaluation of Public Output: 38549 Public Choice and the Benefit Principle of Taxation WPS1091 Capital Market Imperfections Before Fidel Jaramilb February 1993 W. Pitayatonakarn and After Financial Liberalization: Fabbo Schiantarelli 37658 A Euler-Equation Approach to Panel Andrew Weiss Data for Ecuadorian Firms WPS1092 The Effect of Financial Liberalization Fidel Jaramillo February 1993 W. Pitayatonakarn on the Allocation of Credit: Panel Fabio Schiantarelli 37658 Data Evidence for Ecuador Andrew Weiss WPS1093 Swiss Chilanpore: The Way Forward Dimitri Vittas February 1993 W. Pitayatonakarn for Pension Reform? 37658 WPS1094 The New Regionalism: A Country Jaime de Mob February 1993 D. Ballantyne Perspective Arvind Panagariya 37947 Dani Rodrik Policy Research Working Paper Series Contact Title Author Date for paper WPS1095 Are Failproof Banking Systems Samuel H. Talley February 1993 B. Lu Feasible? Desirable 37664 WPS1096 How Trade Liberalization Affected Mona Haddad February 1993 D. Ballantyne Productivity in Morocco 37947 WPS1097 A Production Function-Based Policy Takamasa Akiyama February 1993 G. llogon Simulation Model of Perennial Jonathan R. Coleman 33732 Commodity Markets WPS1098 Fortress Europe and Other Myths Jean Baneth February 1993 M. St. Leger Concerning Trade 37148 WPS1099 Trade Unions and Collective Harry C. Katz February 1993 ESP Bargaining Sarosh Kuruvilla 33680 Lowell Turner