|
You are here: Eurofound > EIROnline > 2004 > 01 > Industrial relations in the EU, Japan and USA, 2002 My Eurofound: Login or Sign Up   

Industrial relations in the EU, Japan and USA, 2002

This overview compares a number of key aspects of industrial relations in the EU, Japan and USA, focusing particularly on developments in 2002.

Growing economic globalisation and competition, along with the European Union's deepening economic integration, mean that the EU is increasingly tending to compare itself in many areas with the world's two other largest economies - Japan and the USA. This type of comparison and 'benchmarking' has become particularly important in the light of the EU's commitment, agreed at the European Council summit in Lisbon in 2000 (EU0004241F), to becoming 'the most competitive and dynamic knowledge-based economy in the world capable of sustained economic growth with more and better jobs and greater social cohesion'. Industrial relations systems and developments play an important part in determining economic, employment and social outcomes and are thus a key area of comparison. More broadly, it is widely thought that there is a distinctive 'European social model', which is seen as differing significantly from the models of governing society, the economy and the labour market found in the USA, Japan and elsewhere. Definitions of what exactly constitutes the European social model differ, but it certainly includes the institutions and processes of industrial relations. International comparison in this area can thus help to make clearer what the European social model is and how, and by how much, it differs from the models which characterise other countries.

In this context, this report: compares some key aspects of industrial relations systems in the EU, Japan and USA; looks at a number of important outcomes and developments in 2002; and examines how several important topical issues are being dealt with in the various systems. It should be noted that in such an exercise the normal problems of international comparison in industrial relations are exacerbated by several factors. First, the comparison is between, on one side, a set of countries which are grouped in a union which has common rules and policies in a number of areas (economic, social, monetary, employment etc) but are still relatively diverse (if integrating to some extent) and, on the other side, two individual countries. More appropriate comparators for the EU might be the North American Free Trade Agreement (NAFTA) countries and the Association of South East Asian Nations (ASEAN). Second, the EU is in the process of enlarging, with 10 new Member States from central and eastern Europe and the Mediterranean joining in May 2004. While the industrial relations systems of the current 15 EU Member States are relatively stable and well researched, those of the accession countries are in most cases evolving rapidly and data is hard to find in some areas. There are thus information gaps for the new Member States, while enlargement makes it more difficult to treat the EU as a single bloc for comparative purposes. With these caveats, this report first examines a number of basic industrial relations actors, structures and processes:

  • trade unions;
  • employer organisations;
  • collective bargaining; and
  • (briefly) employee involvement and employment legislation.

The report then goes on to look at the economic context for industrial relations in 2002 and pay and working time outcomes in that year, before examining how the various industrial relations systems deal with two key topical issues:

  • company restructuring; and
  • 'new forms of work'.

The information in this report on the EU is taken mainly from the European Industrial Relations Observatory (EIRO), while the data for Japan and the USA are taken largely from special reports on industrial relations developments in 2002 prepared for EIRO by experts in these countries - see JP0311101F and US0311101F- plus sources such as the US Bureau of Labor Statistics (BLS) and the Japan Institute for Labour Policy and Training (JILPT). Furthermore, parts of the report are informed by the discussions at a conference on 'Industrial relations in the EU, USA and Japan' organised in Dublin on 6-7 November 2003 by the European Foundation for the Improvement of Living and Working Conditions and attended by experts from the countries concerned and international organisations. Information on industrial relations developments in the EU, Japan and USA in 2000 and 2001 and background information on other aspects of their systems is provided in TN0111148F and TN0212101F.

Trade unions

Organisation

There is a varied pattern of trade union organisation in the current EU Member States, and it is hard to discern any typical 'EU model' of union structure. The simplest situation is found in Austria and Germany, where there is essentially a single confederation (with only relatively minor alternative organisations in Germany, and none in Austria) made up of a quite small number of primarily industrial unions. There is also a single major national centre in Ireland and the UK, though with greater numbers of affiliates, which are a mixture of industrial, general and occupational unions.

In Belgium, France, Italy, Luxembourg, the Netherlands, Portugal and Spain, there are multiple competing union confederations, divided (at least originally) mainly on political and religious grounds - with the number of main or most 'representative' confederations varying from five in France to two in Portugal and Spain. In many cases, these confederations are made up of sectoral federations. In some of these countries - notably France and Italy - there are significant trade unions outside these 'representative' confederations. In the Nordic countries, the general picture is of separate confederations for different occupational groups - typically blue-collar, white-collar and professional/academic - made up of industrial and occupational unions. Finally, the distinction in Greece is between two confederations (made up of a variety of types of member organisation) for the private and public sectors (though there are moves towards a merger between the two organisations).

A similar diversity of trade union organisation is found in the 10 new Member States. The Czech Republic, Estonia, Latvia, Slovakia and Slovenia have essentially a single dominant confederation, with smaller rivals of varying size (except in Latvia). In the other five countries, there are multiple competitive confederations, divided (at least originally) mainly on broadly political grounds though sometimes on occupational grounds. The number of main centres varies from two in Malta and Poland to six in Hungary. The divisions between unions in many Central and Eastern European (CEE) new Member States arise, in varying ways, from the distinctions between: new unions created after the fall of the old political system in the late 1980s and early 1990s; and former state-dominated unions which have reformed themselves in the new economic and political context (and in some cases split into occupational unions, often divided between private and public sectors, as in Hungary).

Despite this diversity and sometimes division at national level, trade unionism in the current and enlarged EU is characterised by a high degree of unity and coherence at European level. The European Trade Union Confederation (ETUC) brings together virtually all major confederations and centres in the current Member States (with a number of gaps in coverage filled in recent years). While coverage in the new Member States is not yet as comprehensive, ETUC has affiliates in all 10, with the largest centres appearing to be members in most countries. ETUC extends beyond the enlarged EU 25 and currently has 77 member organisations from 35 countries, with a total membership of around 60 million. Also affiliated to ETUC are 11 'European industry federations', grouping almost all major EU trade unions in their respective sectors, along with many from the new Member States.

The most notable exceptions to ETUC membership among union organisations in the current EU are: a number of specific organisations for managerial and professional staff, which belong to the European Confederation of Executives and Managerial Staff (CEC); and a number of organisations (generally outside the trade union 'mainstream') affiliated to the European Confederation of Independent Trade Unions (CESI). Both CEC and CESI have some full or observer members in the new Member States.

The structure of trade unionism in the USA is relatively straightforward and similar to that in Ireland and the UK. There is a single main national centre, the American Federation of Labor and Congress of Industrial Organisations (AFL-CIO), made up of a relatively large number (currently 64) of industrial and occupational unions. Only a few unions lie outside the AFL-CIO umbrella.

In Japan, there are two confederations of significant size. By far the larger is the Japanese Trade Union Confederation (Rengo), which organises some 64% of unionised workers. Rengo's membership is based largely on enterprise-level unions, organised in sectoral federations (there are thus some similarities with a country like France, where the main union confederations are made up essentially of local unions organised in sectoral federations). The second confederation, the National Confederation of Trade Unions (Zenroren), represents only around 9% of all unionised workers.

A common trend across the EU (at both European and national levels), Japan and the USA is towards consolidation and merger of trade union organisations. The number of European industry federations affiliated to ETUC has been reduced by mergers in recent years (eg in the food and agricultural sectors, and in services), while the number of member unions of most national union confederations has declined (and there are plans for further rationalisation in some countries). For example, 2002-3 saw major union mergers initiated or completed in countries such as Austria, Denmark, Sweden and the UK, while the two main Greek confederations are considering merger (GR0212102F). This trend appears less pronounced in the new CEE Member States, where the period of economic and political transition was in many cases marked - especially in the early years - by the emergence of numerous competing union centres and splits from existing centres. There have recently been some moves towards a greater rationalisation of union structures in a number of countries, such as Hungary, though conversely a new national centre was set up in Poland in 2002, alongside the two existing ones (PL0212109F).

In the USA, the number of unions affiliated to AFL-CIO has fallen from 96 in 1985 to 64 today. However, many of these unions are still small, and the 15 largest represent over 75% of total AFL-CIO membership between them. In 2002, a number of leading figures in the union movement launched proposals to consolidate union membership into 10-15 large unions that would each focus on specific sectors, industries and labour markets. Several major unions - the Service Employees International Union (SEIU), the Hotel Employees and Restaurant Employees International Union (HERE), the Union of Needletrades, Industrial and Textile Employees (UNITE), the Laborers' International Union of North America (LIUNA) and the United Brotherhood of Carpenters (UBC) - subsequently endorsed this consolidation initiative, known as the New Unity Partnership (NUP). It is not yet clear if the NUP will gain sufficient support within AFL-CIO, especially from the leaders of the numerous small unions which might be subsumed by larger ones. It has been suggested that the NUP’s sponsors, if obstructed, could take unilateral action and set up a new rival federation to AFL-CIO.

A process of consolidation has been occurring among federations affiliated to Japan's Rengo, albeit a relatively slow one. The most significant merger for some years occurred in 2002, when a new industrial trade union called UI Zensen was formed by the fusion of: Zensen, an industry union covering mainly the textiles and distribution sectors (with 621,733 members); the Japanese Federation of Chemical, Service and General Trade Unions (CSG Rengo) (171,781); and the small Federation of Textile, Clothing and Living Goods Workers' Unions of Japan (Sen’i-seikatsu-roren). The new organisation has a membership of 795,000, making it the largest industrial union in the private sector, representing workers at 1,989 companies.

Membership and density

Overall, union membership has generally been falling across the industrialised market economies in recent decades - according to most commentators, this is due largely to common trends such as a decline in employment in traditionally high-unionisation manufacturing industry and the growth of lower-unionisation services employment, and increasing levels of 'atypical' employment. This decline seems to have occurred relatively uniformly across the EU, Japan and the USA - though there are some exceptions in the current EU (eg Denmark, Finland and Ireland), and the decline has been relatively minor or recent in some cases. To take the example of Germany, in 2002 the total membership of the trade unions affiliated to the German Federation of Trade Unions (Deutscher Gewerkschaftsbund, DGB) - the country's principal union centre - fell by 2.5% compared with 2001 (DE0302201N). However, this was the first time in a decade that DGB’s affiliates had lost less than 3% of their members.

In 2002, union membership declined by 3.7% in Japan, following falls of 2.8% in 2001 and 2.4% in 2000. These figures confirm a continuing, gradual decline in union membership over the long term. In the USA, union membership fell by 1.7% in 2002, following a rise of 0.8% in 2001 and a fall of 1.3% in 2000. No such overall figures are available for the EU, but decline or at best stability seems to have been the overall picture in almost all cases for the past few years.

The new CEE Member States are something of a special case, with trade unions in many countries experiencing very major falls in membership over the period of economic and political transition. This is especially true in countries such as the Baltic states, Poland and Hungary, though less so in Slovakia and Slovenia. Given that union membership was in many cases compulsory under previous communist regimes, a fall of some kind was to be expected with the introduction of democracy, but additional factors which have been identified as contributing to union decline (TN0207104F) include: falling living standards over the 1990s; high levels of unemployment; privatisation; growing numbers of small and medium-sized enterprises (SMEs); and sectoral shifts.

Union density - the proportion of those in employment who are union members - is a problematic area of international comparison, due to differing definitions. This applies to how both the total number of trade union members and the total number of potential members are calculated. For example, in some countries, unions have many members who are retired, unemployed, students etc and thus not actually in employment (over 20% of union members in Finland fall into this category, for instance). On the other hand, not all employees may be entitled to join unions (eg members of the armed forces, senior civil servants etc). A further problem is how to measure union membership - eg using the unions' own figures or labour market surveys etc. How such issues are dealt with varies from country to country, making international comparisons of density figures very difficult. These problems should be borne in mind when comparing the density figures given below, which are based generally on national definitions which may not be compatible.

Union density has declined in Japan, the USA and almost all European countries in recent years. In Japan, the unionisation rate has been falling steadily since the mid-1970s, when it stood at around 35%. In recent years, it has declined from 22.2% in 1999 to 21.5% in 2000 and 20.7% in 2001, standing at 20.2% in 2002. In the USA, union density stood at 20.1% in 1983 (the first year for which comparable data are available), and has since fallen to 13.9% in 1999, 13.5% in 2000, 13.4% in 2001 and 13.2% in 2002. No overall figures on density trends are available for the current and enlarged EU, but decline is the general picture. For example, in highly unionised Finland, union density fell by 7.3 percentage points over 1994-2001, from 78.5% to 71.2% (despite a slight rise in union membership over the period) (FI0302204F). In Estonia, one of the new Member States where decline has been most dramatic, union density stood at 88.0% in 1992, before collapsing to 37.0% in 1994, since when it has fallen to 26.5% in 1996, 17.4% in 1998, 17.2% in 2000 and 14.2% in 2002 (EE0308101F). Even where union membership losses have been stemmed, increasing employment levels in many countries have meant that union density has fallen - examples include Denmark, Finland and Ireland. Table 1 below gives the latest union density figures available for Japan, the USA and the current and new EU Member States (see above for the caveats which should be borne in mind when comparing figures).

Table 1. Trade union density, Europe, Japan and USA
Country Union density (%)
Denmark 88
Sweden 79
Finland 71
Cyprus 70
Belgium 69
Malta 65
Luxembourg 50
Ireland 45
Unweighted average of 15 current EU Member States 43
Slovenia 41
Austria 40
Unweighted average of expanded EU 25 39
Italy 35
Slovakia 35
Unweighted average of 10 new EU Member States 34
Greece 33
Weighted average of 15 current EU Member States 30
Czech Republic 30
Latvia 30
Portugal 30
Germany 30
Weighted average of expanded EU 25 29
UK 29
Netherlands 25
Weighted average of 10 new EU Member States 22
Hungary 20
Japan 20
Lithuania 15
Poland 15
Spain 15
Estonia 14
USA 13
France 9

Source: 2002 figure for Japan from JILPT; 2002 figure for USA from BLS; figures for individual European countries are the latest reported in EIRO, generally covering 2000-2; weighted averages produced by applying reported density rates to total employee figures as recorded in the Eurostat Labour Force Survey 2002.

As the table indicates, the average trade union density in the current EU (unweighted for the different sizes of the 15 countries) is more than double that in Japan and more than three times that in the USA. Union membership is obviously an altogether more common phenomenon in Europe than the clearly minority activity it is in Japan and the USA. However, the unweighted average for the present EU does not represent the proportion of the total EU workforce who are union members. Density in the largest countries - France, Germany, Italy, Spain and the UK - is considerably lower than the unweighted average (very much so in the cases of France and Spain), so the true average is substantially less. Weighting the density figures in line with the relative size of the countries' workforces gives an average for the current EU of 30%. Union density varies considerably among EU Member States, from around 70% or more in Belgium, Denmark, Finland and Sweden to 15% or lower in France and Spain. However, the USA's union density is lower than any EU country's except France, while Japan's density exceeds only that of France and Spain.

Union density is generally lower in the new EU Member States than in the 'old' 15, with their unweighted average around a quarter lower than that for the current EU. Weighting the two averages gives a current EU figure which is nearly half as high again as the figure for the new entrants. This reflects the fact that union density is below average in the largest acceding countries - Poland, the Czech Republic and Hungary - though relatively high in smaller countries such as Cyprus and Malta. The expansion of the EU brings the unweighted average union density down from 43% to 39% but cuts the unweighted average by only one one point from 30% to 29%. Expansion will therefore bring the overall EU density situation somewhat nearer to Japanese and US levels, but only slightly.

In gender terms, unionisation levels are higher among men than women in the USA and probably Japan (though no figures are available in the latter case). This reflects the pattern in many EU countries with low to medium overall unionisation rates - such as Austria, Germany, the Netherlands and the UK. However in the high-unionisation Nordic countries - Denmark, Finland and Sweden - women's unionisation rate exceeds that of men (TN0103201U).

Employer organisations

In the current EU, the organisation of employers varies considerably between the Member States. At national intersectoral level, in countries such as Belgium, Denmark, France, Ireland, Italy, Luxembourg, the Netherlands, Spain, Sweden and the UK, there is essentially a single umbrella organisation (at least for the private sector) representing companies' employer and business/trade interests, though accompanied by separate SME organisations in some cases (as in France, the Netherlands or Spain). In other countries - notably Germany - there is a division between the representation of employer and of business/trade interests, with separate central organisations for each. The trend, however, appears to be towards the unification of representation of employer and business/trade interests. Another difference at intersectoral level is that there may be a single central (private sector) body - as in the countries mentioned above - or there may be separate bodies for industry, services and in some cases agriculture - as in Finland, Greece and Portugal. Again, in some countries there are moves to merge such separate organisations representing broad areas of the economy, as in Finland (FI0310201N).

Turning to the new Member States, while Cyprus and Malta have long-standing central employer bodies, employer organisations in the genuine sense have emerged in the CEE countries only since the economic and political transformation of the late 1980s and early 1990s. Initially, these organisation mainly represented state-owned enterprises, but privatisation has since resulted in a proliferation of employer and industry organisations in some countries, especially for SMEs. Many organisations have essentially a business/trade role, focusing on economic issues and lobbying, rather than employer issues and collective bargaining (though many participate in various national-level tripartite forums, which may involve engagement in quasi-bargaining activities and in shaping public policy on employment and wider social and economic issues). A lack of strong and representative employer organisations at national and sectoral level is a frequently noted feature of many CEE countries.

There is essentially a single main organisation in Cyprus, Estonia, Latvia, Lithuania and Slovakia. Slovenia is unusual (SI0207103F) in that (rather like Austria) it has 'chambers', of which membership is obligatory for employers - though alongside these compulsory bodies, there are separate employer organisations with voluntary membership. The Czech Republic and Malta have two main central bodies, while Poland and Hungary have four or more.

In terms of the role of employer organisations, in the current EU regular national intersectoral bargaining with trade unions over substantive pay and conditions issues is part of the remit of central employer bodies in Belgium, Finland, Greece, Ireland and Portugal. Intersectoral bargaining over specific issues or procedural matters is part of the employer confederations' role in Denmark, France, Italy, Spain and Sweden. While usually falling short of bargaining, employer confederations have close cooperative relations with trade unions in various fora in Austria, Germany and the Netherlands, which may lead to joint texts or approaches. It is perhaps in the UK that the main employer body (the Confederation of British Industry) has the least 'bargaining-like' role in any area.

In the new Member States, in almost all cases employer organisations do not engage in bipartite national intersectoral bargaining with trade unions over pay and conditions - the main exceptions being Slovenia and Latvia (on minimum wages). Instead, there is a high degree of tripartism at national level, with employer bodies involved in negotiation and/or consultation processes with trade unions and governments. This may result in regular tripartite agreements on minimum wages, as in Hungary, or wider issues, as in the Czech Republic and Slovakia. Less regular or issue-specific tripartite national agreements have been concluded in Cyprus, Estonia, Latvia, Lithuania, Malta and Poland. Slovenia has tripartite national pay agreements, as well as bipartite national bargaining.

At industry level, sectoral employer organisations with a collective bargaining role are key components of the industrial relations systems of most current EU countries - Austria, Belgium, Denmark, Finland, France, Germany, Greece, Italy, Netherlands, Portugal, Spain and Sweden. Only in Ireland, Luxembourg and the UK are there few sectoral employer associations with a bargaining role, as collective bargaining occurs essentially at company level (though overlaid with intersectoral bargaining in Ireland - see below under 'Collective bargaining').

In the new Member States, sectoral employer organisations have much less of a bargaining role, with the notable exceptions of Slovenia, Cyprus and, to a lesser extent, Slovakia. This is due to the fact that in most of these countries sectoral employer organisations are either weak and lack the necessary resources to participate and/or do not have the authority to conclude sectoral agreements on behalf of their members in those cases where they are better established - eg in Hungary and in Poland.

As with trade unionism, the representation of employers' interests at EU level is relatively comprehensive and coherent. The Union of Industrial and Employers' Confederations of Europe (UNICE) represents almost all the main national intersectoral confederations of private sector employers and business in the current EU Member States. It acts as both an employers' organisation (in that it engages in dialogue and, in specific circumstances, negotiations with ETUC) and as a trade/industry association (in that it is involved in promoting its members' interests in a range of areas, and in seeking to influence EU decision-making in areas of relevance). UNICE's coverage of organisations representing SMEs is arguably patchy, and a separate European-level body, the European Association of Craft and Small and Medium-sized Enterprises (UEAPME), seeks to represent this category of businesses. Since 1998, UEAPME and UNICE have cooperated closely in EU-level social dialogue and negotiations with ETUC (EU9903159F). Furthermore, the European Centre of Enterprises with Public Participation and of Enterprises of General Economic Interest (CEEP) represents enterprises and organisations with public participation or carrying out activities of general economic interest, whatever their legal or ownership status. It is treated as central social partner organisation alongside UNICE by the European Commission, and is involved in dialogue and negotiations with ETUC.

UNICE - which currently has a total of 35 national member organisations and four observers - has member organisations in seven of the 10 new Member States (Cyprus, the Czech Republic, Estonia, Hungary, Malta, Poland and Slovakia) and observer members in Lithuania, Poland and Slovenia, though apparently no affiliate in Latvia. UEAPME has full members only in the current EU Member States, but associate members in all 10 new Member States. CEEP also has full members only in the current Member States, and associate members only in Hungary among the new Member States

At European sectoral level, there are hundreds of organisations representing business interests. However, very few of these are employer organisations, in the sense that they represent their members on employment issues or have relations with trade union organisations. The main exceptions are the organisations in those sectors where a 'sectoral social dialogue' has developed, either autonomously or on the instigation of the European Commission. There are currently around 30 sectoral dialogue committees, in most of which joint texts (opinions, declarations, codes of conduct etc) on a range of issues (eg training, employment, fundamental rights or health and safety) have been reached. However, the employer bodies (and union organisations) do not have a genuine bargaining role over pay and conditions, except in exceptional circumstances. The sectoral employer organisations involved in the dialogue generally represent all current EU Member States, though their membership in the new Member States varies considerably.

Until 2002, Japan was like a number of European countries in having a single central employers' body - the Japan Federation of Employers' Associations (Nikkeiren) - and a separate central organisation representing companies' trade/business interests - the Japan Federation of Economic Organisations (Keidanren). However, reflecting the trend noted above for Europe, Nikkeiren and Keidanren merged in 2002 to form the Japan Business Federation (Nippon Keidanren). Nippon Keidanren's membership is made up of 1,268 individual companies, plus 126 industrial associations and 47 regional employer associations. Nippon Keidanren does not participate in collective bargaining with trade unions, but it is involved in formal tripartite dialogue with government and unions, and less formal bipartite dialogue with unions. Furthermore, it attempts to influence the annual 'spring offensive' (Shunto) bargaining round, by issuing guidelines to employers. Its member associations do not generally have any direct bargaining role (though it has been suggested that they play a 'behind the scenes' role in coordinating member companies). Almost all bargaining occurs at the level of the individual enterprise.

The USA is unlike Japan and all EU Member States in that it has no identifiable national intersectoral employer body with an industrial relations role. A special organisation has been created to allow US employers to be represented by a single intersectoral body in international organisations and fora - the United States Council for International Business (USCIB). Major business organisations such as the National Association of Manufacturers (NAM) and the US Chamber of Commerce do not deal with trade unions, though they do have some role in developing policy on labour issues. There are few national sectoral employer bodies with any bargaining role, and bargaining takes place predominantly at enterprise or local level (see below under 'Collective bargaining').

The organisation rate or 'density' of employer organisations is hard to assess, due to frequent lack of data and difficulties of definition, and often somewhat contested. No data are available for Japan and the issue is not really relevant in the USA. The data available from current and future EU countries are patchy and hard to compare. It can be said (drawing on European Commission and EIRO research) that among countries for which data are available, the proportion of the total national workforce employed by the members of the main central employer organisations is almost total in Austria (where membership of the Chamber of the Economy [Wirtschaftskammer Österreich, WKÖ] is compulsory), relatively high in Luxembourg, the Netherlands, Germany, Finland, Ireland, Italy and France, and somewhat lower, while still significant, in Denmark and the UK. In most cases, membership levels seem quite stable, though a decline has been witnessed in a few countries such as Germany. Data are even scarcer for the new Member States, but it seems that density is generally likely to be lower. The European Commission has estimated the membership of employer organisations in the candidate countries at an average of 30%-40% of industrial enterprises and 2%-5% of all enterprises. Countries known to have higher levels of membership include Slovenia, where chamber membership is compulsory for employers and Cyprus, where member companies of the Employers and Industrialists Federation (OEB) employ some 57% of the total labour force.

Collective bargaining

Collective bargaining plays a key role in industrial relations in all current EU Member States, though national systems differ very widely in terms of the level, coverage, content and nature of bargaining. In the new Member States, collective bargaining generally remains a rather weak institution, though with some exceptions. On the whole, in terms of the coverage of bargaining and the number of agreements concluded, collective bargaining in the acceding countries is considerably less developed than in current Member States. In Japan and the USA, the situation is arguably closer to that in many new EU Member States, with bargaining essentially a relatively marginal activity, though it may have a wider impact beyond its direct sphere, especially in Japan.

Bargaining levels

The level at which collective bargaining is conducted is one of the most basic differences in industrial relations between most current EU Member States on one side, and Japan and the USA on the other. In general, bargaining in the present EU is considerably more centralised than in its two competitor countries. However, the new Member States tend to have considerably less centralised systems and their accession will lower the 'average' bargaining level in the Union.

Looking first at Europe, it should be noted that 'bargaining' of a kind occurs at supranational EU level - not something that happens at the level of NAFTA or ASEAN. ETUC, UNICE and CEEP (plus, more recently, UEAPME and the EUROCADRES/CEC liaison committee) have negotiated a number of European agreements, primarily to replace proposed EU legislation in the employment and social field. They have reached such agreements on parental leave (TN9801201S), part-time work (EU9706131F) and fixed-term work (EU9901147F), which have been implemented in law through EU Directives. In July 2002, a framework agreement on teleworking was concluded (EU0207204F), which is being implemented by the members of the signatory parties, rather than by means of a Directive. These various agreements have focused primarily on the current EU Member States. However, the Directives implementing the first three agreements form part of the 'acquis communautaire' which are being adopted and implemented by the new Member States, while the signatories to the telework accord invited their members in the acceding countries to implement the deal. A similar European-level 'bargaining' process driven by the prospect of EU legislation has occurred on some occasions over specific issues in a number of sectors (eg working time in civil aviation and maritime transport), while outside this process numerous joint texts of varying kinds have been concluded by the EU-level sectoral social partners (see above under 'Employer organisations'). As with intersectoral agreements, the primary focus has so far been on the current Member States. However, efforts are being made in the sectoral social dialogue to involve the new Member States more as their accession approaches.

Despite the social dialogue process, concrete pay and conditions of employment are not subject to collective agreement at European intersectoral or sectoral level. There are also no known cases of such concrete bargaining at European level within European companies (though a small but growing number of European Works Councils have reached some form of agreement on specific issues, while cross-border comparisons are increasingly deployed by management and unions in national bargaining, and unions are actively seeking to enhance transnational coordination of bargaining - TN9907201S).

The present EU Member States have widely differing systems of collective bargaining over pay and conditions. However, despite these differences and a widely-observed trend towards the decentralisation of bargaining, most countries have relatively centralised systems. To take the key issue of pay determination, table 2 below indicates the levels at which wage bargaining is currently conducted in the present and new EU Member States, plus Japan and the USA (bargaining over other issues, such as working time or employment, often follows the level of pay bargaining, but may occur at other levels in some countries). It should be noted that the significance of bargaining levels can be subject to change and is not always easy to assess (for example, the relative importance of wage bargaining levels in France is subject to considerable debate).

As table 2 indicates, in three countries (Belgium, Finland and Ireland), the intersectoral level is currently the dominant wage bargaining level, while some bargaining at this level also occurs in Denmark, Greece (minimum pay rates), the Netherlands (where recommended pay increases are sometimes agreed at this level, as for 2003-4 - NL0310103F) and Spain (where the central social partners have agreed pay recommendations for lower-level bargaining in recent years - ES0302204F). In eight countries (Austria, Germany, Greece, Italy, the Netherlands, Portugal, Spain and Sweden) the sector is the most important level of wage bargaining, while some bargaining occurs at this level in all the other Member States. In three countries there is no predominant bargaining level (Denmark, France and Luxembourg). In only the UK is the company the key pay bargaining level, while it is also important in France and Luxembourg and exists in all the other Member States.

Although there is not a clear-cut distinction between the two groups of countries - eg the UK has a very decentralised bargaining system and Slovenia a very centralised one - collective bargaining is generally much more decentralised in the new Member States than is the case in the current EU. The company is the dominant pay bargaining level (though, given low bargaining coverage rates in many countries - see below - this does not necessarily mean that it covers a high proportion of all companies and employees) in all 10 countries apart from Slovakia, Slovenia and Cyprus. Intersectoral bargaining is absent from all countries apart from Slovenia and, to a lesser extent, Hungary and Latvia. Sectoral bargaining (in a broad sense) plays the dominant role only in Slovakia and Cyprus, is a very important bargaining level in Slovenia, and a relatively significant bargaining level in Hungary and to a lesser extent the Czech Republic. It exists in Estonia, Latvia and Poland, but is of very limited significance. The impact of EU enlargement will arguably be to shift the 'average' level of bargaining towards the company level. In only one of the current 15 Member States is the company the most important pay bargaining level, but in the expanded EU of 25 there will be eight such countries (ie nearly a third of the total). A key role for sectoral bargaining is often seen as a defining characteristic of the'EU model' of industrial relations, and of the 15 current Member States, it is a dominant or important level of pay bargaining in 11 cases (ie around three-quarters of the total). However, in the EU of 25, this will be true of only 14 countries (under 60% of the total).

In both Japan and the USA, the predominant bargaining level for pay and all other issues is the individual company. Intersectoral or sectoral bargaining is virtually unknown (with some exceptions at industry level in the USA). As noted above, this is also the case (for pay) in the UK and seven out of 10 of the new Member States. However, the intersectoral or sectoral level is still the dominant or most important level of wage bargaining in at least 15 of the 25 countries (60%) of the extended EU and an important level in two more. These countries include four of the five largest EU economies (France, Germany, Italy and Spain). It is thus clear that collective bargaining is much more centralised in the EU than is the case in its two major competitors.

However, it should be noted that in Japan, there is a degree of coordination on both trade union and employers' sides during the annual 'Shunto' bargaining round. The general pattern for these spring-time negotiations is that major manufacturers in leading industries such as electrical goods or cars take the lead in bargaining and are then followed by other large companies and then by SMEs.

Collective bargaining in the USA has become increasingly decentralised in recent decades. Post-war 'pattern bargaining' arrangements (whereby an agreement is achieved in a particular company or companies and then extended to other firms in the sector) have largely unravelled over the last 25 years and multi-employer 'master' agreements covering sectors such as trucking, basic steel, tyres, airlines and coal mining have collapsed under the weight of deregulatory forces and pressures on unions to make local concessions on pay and other issues. The focus of bargaining has thus shifted over time, first from multi-employer pattern bargaining to the company level, and then from the company level to the individual worksite.

Table 2. Wage bargaining levels in Europe, Japan and the USA
. Intersectoral level Sectoral level Company level
Austria . XXX X
Belgium XXX X X
Cyprus . XXX X
Czech Republic . X XXX
Denmark XX XX X
Estonia . X XXX
Finland XXX X X
France . XX XX
Germany . XXX X
Greece X XXX X
Hungary X XX XXX
Ireland XXX X X
Italy . XXX X
Japan . . XXX
Latvia X X XXX
Lithuania . X XXX
Luxembourg . XX XX
Malta . . XXX
Netherlands X XXX X
Poland . X XXX
Portugal . XXX X
Slovakia . XXX X
Slovenia XXX XX X
Spain X XXX X
Sweden . XXX X
UK . X XXX
USA . X XXX

X = existing level of wage bargaining; XX = important, but not dominant level of wage bargaining; XXX = dominant level of wage bargaining.

Source: for current EU Member States, a 2000 EIRO study on pay setting ( TN0007402S ), with author's updates/adjustments; for new EU Member States, a 2002 EIRO study on industrial relations in the EU and candidate countries ( TN0207104F ); for Japan and USA, author's assessment.

Bargaining coverage

The coverage of collective bargaining ie the proportion of workers that have their pay and conditions set, at least to some extent, by collective agreements is generally high (and relatively stable) in the current EU Member States - see table 3 below (no data are available for Greece or Malta). The coverage rate is two-thirds or more in all 14 countries for which information is available except Luxembourg and the UK (where only a little over a third of employees have their pay set by collective bargaining), and is as high as 90% or more in France, Belgium, Sweden, Finland and Italy. High bargaining coverage rates are found in most countries with a system of sectoral collective agreements, such as Austria, Denmark, France, Germany, Italy, the Netherlands, Spain and Sweden. In some cases - such as Austria, France, Germany and the Netherlands - systems of extending the binding contents of sectoral collective agreements to employers and employees that are not members of signatory organisations also contribute to high levels of bargaining coverage. In Finland, Greece and Ireland, high levels of bargaining coverage are achieved by intersectoral agreements. The lowest coverage rates are found in the countries where bargaining is most decentralised - the UK and Luxembourg. The average bargaining coverage rate in the current EU (except Greece) - unweighted for the different sizes of the 14 countries concerned - stands at around 77%. Weighting the density figures in line with the relative size of the countries' workforces produces an average of 72%.

In general, bargaining coverage is lower in the new CEE Member States, mirroring the generally decentralised level of bargaining and lower levels of trade union membership. Slovenia is the main exception, with a centralised bargaining system ensuring almost total bargaining coverage. Elsewhere, bargaining seems to cover a minority of the workforce, though relatively high coverage rates (if still low in EU terms) are also found in Slovakia and Hungary, where sectoral bargaining is more significant than in most CEE countries. Although extension procedures exist in a number of countries, these appear to have little impact, with sectoral agreements rare. Bargaining coverage in Cyprus is around two-thirds and, while exact figures are not available for Malta, it appears that coverage here too is higher than the norm in the CEE countries. Taking the nine new Member States for which data are available (ie excluding Malta), the unweighted average bargaining coverage rate is 42% and the weighted average 37% - both figures are only a little over half of the rates in the current EU. The accession of the generally low bargaining coverage new Member States will reduce the EU's overall unweighted coverage rate substantially, from 77% to 63%, but will cut the weighted average only from 72% to 67%.

Something like two-thirds of the workforce of the expanded EU are thus covered by collective bargaining - a rate which is over three times that found in Japan and over four times that in the USA. In Japan, the fact that bargaining is conducted exclusively at company level, with no mechanisms for the extension of agreements beyond the signatories, means that bargaining coverage exactly matches trade union density, at a little over 20%. However, the wage increases agreed in the 'bargaining sector' have a major influence on general wage levels, both setting the market rate in the 'non-bargaining sector' and being passed on through recommendations on the wages of public servants and through the minimum wage system. In the USA, the coverage of bargaining (essentially conducted at company level) is similarly restricted to trade union members - ie 13.2% of the workforce (16.1 million people) in 2002. However, to this figure can be added a further 1.7 million workers organised in a 'union-typical' manner, such as members of certain employee associations, or non-union members whose jobs are covered by a union or employee association contract, bringing the number of workers represented at the bargaining table to 17.8 million, or 14.6% of the total.

Bargaining coverage in the current EU seems relatively stable, having declined significantly in recent years only in Germany and the UK and risen only in Denmark (TN0212102S). The picture is more mixed in the new CEE Member States, with examples of both stability and decline. In Japan and the USA, the picture is one of more or less constant decline over the past 20 years or more.

Finally, it should be noted that the bargaining coverage figures used above and in table 3 below are unadjusted rates - ie, they express the number of employees covered by a collective agreement as a proportion of the total number of employees in a country (regardless of whether certain groups are excluded from bargaining). However, in some countries relatively large groups of employees may be excluded from the right to bargain - often public sector employees or certain groups of them (such as the police and armed forces). Excluding these groups from the bargaining coverage figures produces an 'adjusted' coverage rate. In the EU, the adjusted rate is significantly higher than the unadjusted rate in countries such as Austria, Luxembourg and Spain. In the USA, statutory bargaining rights apply to only 78% of workers, with excluded groups including: supervisors, managers and independent contractors; public sector workers in 12 states; domestic workers; and many agricultural workers. The adjusted coverage rate thus stands at around 18.7%, around four percentage points higher than the unadjusted rate.

Table 3. Direct collective bargaining coverage, Europe, Japan and Europe
Country Coverage
Slovenia 100%
France 90%-95%
Belgium 90%
Sweden 90%
Finland 90%
Italy 90%
Netherlands 88%
Portugal 87%
Denmark 83%
Austria 78%
Unweighted average of 14 current EU Member States* 77%
Weighted average of 14 current EU Member States 72%
Spain 68%
Germany 67%
Weighted average of expanded EU 23* 67%
Ireland 66%
Cyprus 65%-70%
Unweighted average of expanded EU 23 63%
Luxembourg 48%
Slovakia 48%
Unweighted average of 9 new EU Member States* 42%
Poland 40%
Weighted average of 9 new EU Member States 37%
UK 36%
Hungary 31%
Estonia 28%
Czech Republic 25%-30%
Japan 20%
Latvia Under 20%
USA 15%
Lithuania 10%-15%

* No data are available for Greece and Malta.

Note: figures represent % of total workforce directly covered by collective bargaining, and are not adjusted to exclude those groups of workers excluded by law from bargaining coverage; weighted averages produced by applying reported coverage rates to total employee figures as recorded in the Eurostat Labour Force Survey 2002.

Source: figures for Europe as calculated (for 2000-1) for 2003 EIRO study on bargaining coverage ( TN0212102S ), except for Cyprus, Czech Republic, Latvia, Lithuania (from a 2002 EIRO study on IR in the candidate countries - TN0207102F ), Estonia, Finland and Italy (from other EIRO records), Ireland (from the European Commission's Employment in Europe 2003, and Poland (from ETUC); figure for Japan (2002) is from JILPT; figure for USA (2002) from BLS.

Employee involvement

A key feature of industrial relations in most EU Member States is the existence of a widespread system of indirect or representational employee involvement at company or workplace level through 'works councils' or similar bodies. Standing bodies of employee representatives (plus management in some countries) elected by employees (sometimes from trade union lists) and set up on the basis of law or collective agreement, with defined rights to information, consultation and sometimes other forms of involvement, exist in 11 of the 15 current Member States: Austria, Belgium, Denmark, France, Germany, Greece, Italy, Luxembourg, the Netherlands, Portugal and Spain. In Finland, legislation provides for exchange of information and 'cooperation' negotiations between employers and employees and/or their representatives, while in Sweden legislation confers information, consultation and co-determination rights on trade unions. All these 13 countries thus have a general system providing for ongoing information and consultation for employee representatives of some type. The exceptions to the rule are Ireland and the UK. Here, trade unions - where recognised - are the primary channel for information and consultation. Statutory information and consultation rights are limited to a number of specific issues (notably collective redundancies and transfers of undertakings) and there are also - especially in the UK - 'joint consultative committees' or similar structures in some companies, based on collective agreements or voluntary practice. This situation is set to change to some extent in these two countries owing to the implementation of the recent EU Directive (2002/14/EC) establishing a general framework for informing and consulting employees (EU0110206F). This Directive provides employees in undertakings or establishments over a certain size with rights to information and consultation on a number of business, employment and change issues. Its implementation will have the greatest effect in Ireland and the UK, and may lead to a greater dissemination of works council-type arrangements in these countries.

Statutory works council-type bodies (with varying relationships to trade union channels of representation) also exist in some of the new Member States - notably the Czech Republic, Hungary, Poland (state-owned companies only), Slovakia and Slovenia. Compliance with the 2002 EU information and consultation Directive is likely to lead to a wider introduction of such structures or similar procedures.

In addition, in a majority of current EU countries, there is a statutory system for some form of employee representation on the board of directors or supervisory boards of some types of company (TN9809201S). Such participation is relatively widespread in Austria, Denmark, Finland, France, Germany, Luxembourg, the Netherlands and Sweden, and restricted to some public sector organisations in Greece and Ireland. A number of acceding countries also have such arrangements - examples include Hungary, Malta, Poland (former state enterprises) and Slovenia.

At EU level, legislation has introduced European Works Councils (EWCs), pan-European structures for the information and consultation of employees and their representatives on a range of business and employment issues in multinational companies over a certain size operating in the EU. Furthermore, as mentioned above, Directive 2002/14/EC provides that employees in all Member States (employed in companies or establishments over a certain size) should be informed and consulted on business, employment and work organisation matters. Specific Directives have guaranteed information and/or consultation on specific issues, notably business transfers, collective redundancies and health and safety.

The situation is very different in the USA. There is no legislation providing for works council-type structures or board-level employee representation. Nevertheless, in 1994 a survey conducted by the Dunlop Commission (see below), found that 75% of large employers had some form of employee involvement. These types of involvement vary considerably, ranging from teams that deal with specific problems for short periods to groups that meet for more extended periods. They include 'direct' forms such as quality circles and self-managed work teams, and not necessarily (or indeed at all) the indirect or representational forms common in the EU. It appears (according to the HR Policy Association, an organisation of senior human resources managers) that many participation arrangements are limited in their scope, or have been curtailed, due to fear of breaches of labour law. This is because the National Labor Relations Act prohibits the formation of an employer-dominated 'labour organisation' or 'employee representation committee' which deals with management over pay, working time or conditions of work. The original purpose of this provision, introduced in 1935, was to prevent employers from controlling company unions. However, as interpreted by the National Labor Relations Board (NLRB), the law results in severe restrictions on the formation of non-union teams or committees that address 'terms and conditions of employment'- which is taken to include a wide range of matters such as pay, training, productivity/efficiency rewards, work assignments, job descriptions/classification, workloads/quotas, recruitment and dismissal. However, a number of recent NLRB decisions (notably the Crown Cork & Seal case in 1998) may indicate some relaxation of this position. The 1995 report of the Dunlop Commission on the 'future of worker-management relations' recommended that this restriction on employee participation programmes in non-union workplaces should be lifted (though with protection for union representation). A Teamwork for Employees and Managers (TEAM) Act was subsequently passed in Congress, which would have allowed teams of employees in non-union workplaces to work with management to address workplace issues of mutual interest. However, the legislation was vetoed by President Bill Clinton.

In Japan, employee participation at enterprise level is not governed by legislation, but there is a high level of employee-management cooperation. According to a 1999 survey, 41.8% of establishments with 30 or more employees had a 'labour-management consultation organisation'- a permanent structure in which labour and management consult on issues related to management, production, working conditions and welfare programmes. Trade unions were represented in around 85% of these establishments, indicating a high level of inter-relation between the two (65% of employee representatives in labour-management consultation organisations were union representatives). Such bodies are more common in larger firms. Furthermore, a 1996 survey indicated that around 64% of companies had some form of 'employee organisation'- a permanent organisation for employees that holds social gatherings and employee meetings. Japanese labour-management consultation organisations deal with a wide range of issues - notably working time, working conditions, health and safety, welfare, childcare, management policies, overtime premia, pensions, lay-offs and redundancies, restructuring, production/sales and training. Participation takes many forms, notably written explanation by management, exchange of opinions, discussion and agreement.

Issues and outcomes

Having examined a number of basic industrial relations structures and processes above, we will now proceed to look at how the key issues of pay and working time have been dealt with recently, focusing on collective bargaining in 2002, and then at the treatment in the EU, Japan and USA of two topical matters - company restructuring and new forms of work. First, however, we will outline the broad economic and employment context in 2002.

Economic and employment context in 2002

As indicated by table 4 below, the economic and labour market setting for industrial relations differed considerably between the current EU, Japan and the USA in 2002. The EU economy started to slow down in 2001 after several years of rising growth and the downward trend continued through 2002 (and into 2003). Growth was higher and more stable (at 2.4% in 2001 and 2.3% in 2002) in the 10 new Member States (the 'acceding 10'), bringing the rate in the enlarged EU 25 to 1.7% in 2001 and 1.2% in 2002. The Japanese economy remained largely stagnant over 2001-2, with growth falling to virtually zero in the second year. By contrast, after the slump of 2001, the US economy started to grow again in 2002, albeit unevenly.

US inflation continued to fall, from 2.8% in 2001 to 1.6% in 2002, while the EU 15 rate eased slightly from 2.3% to 2.1%. Inflation in the acceding 10 was much higher at 5.1% in 2001, but fell to 2.5% in 2002, while the enlarged EU 25 had a rate of 2.4% in 2001 and 2.1% in 2002. Continuing the trend of recent years, prices continued to fall in Japan.

Employment levels rose slightly in the EU 15 during 2001 and by even less in 2002. However, while small, these increases contrasted with the negative trend in Japan and the USA. After many years of growth, employment fell slightly in the USA in 2001 and 2002, while greater falls were recorded in Japan, where employment has been declining since the late 1990s. Nevertheless, EU employment rates are still some way below those in both Japan and the USA - a gap that will grow when the new Member States (which have an average employment rate nearly 10 percentage points below that in the current Member States) join the EU. In all three cases, male employment rates exceed female rates.

Unemployment rose in the EU, Japan and the USA in 2002. The US unemployment rate increased to 4.8% in 2001 following many years of decline, and then rose to 5.8% in 2002. Japan saw a continuation of a long-term upward trend, with the rate in 2002 standing at 5.4%. EU 15 unemployment rose slightly between 2001 and 2002, ending a downward trend. However, the EU 15 unemployment rate remained about half as high again as that in both Japan and the USA. Again, this gap will increase following EU enlargement. The unemployment rate in the acceding 10 is nearly double that in the current EU, bringing the overall rate in the enlarged EU 25 to 8.6% in 2001 and 8.9% in 2002.

Nominal pay increases per employee in the EU 15 exceeded those in Japan and the USA in both 2001 and 2002 (increases in the acceding 10 were higher still, at 11.2% in 2001). The rate of increase fell between 2001 and 2002 in both the EU 15 and USA, while in Japan, where nominal pay has been falling for several years, the rate of decline sharpened. Adjusting nominal compensation increases for inflation, in 2001 real increases were greatest in the EU 15, followed by Japan and the USA, while in 2002, the USA took over the leading position. The rate of increase in labour productivity rose in all three cases from 2001 to 2002, with the USA achieving the greatest rise. In 2002, US labour productivity growth, at 3.1%, was double that in Japan and four times that in the EU 15 (productivity was growing faster in the acceding 10, at 2.5% in 2001 and 3.1% in 2002). In 2001, nominal unit labour costs grew somewhat faster in the EU 15 than in the USA (and by as much as 8.5% in the acceding 10), while they fell in Japan. In 2002, the rate of increase slowed somewhat in the EU 15, but dropped more dramatically in the USA, with costs actually falling by 0.5%. Japan's longer-established downward trend deepened further. Finally, adjusting the nominal unit labour cost increases for inflation, increases were greatest in the EU in both 2001 and 2002, followed by Japan and the USA.

Table 4. Economic and employment indicators, EU 15, Japan and USA, 2001-2
. EU 15 Japan USA
. 2001 2002 2001 2002 2001 2002
GDP growth 1.6 1.1 0.4 0.2 0.3 2.5
Inflation (CPI) 2.3 2.1 -0.6 -0.9 2.8 1.6
Employment rate ('benchmark') 65.2 65.2 76.9 76.3 81.3 80.3
Employment growth 1.4 0.5 -0.6 -1.2 -0.3 -0.3
Unemployment rate (Eurostat definition) 7.4 7.7 5.0 5.4 4.8 5.8
Growth in nominal compensation per employee 3.5 3.0 -0.4 -1.7 2.9 2.6
Growth in real compensation per employee (private consumption deflator) 1.4 0.9 1.2 -0.2 0.9 1.2
Labour productivity growth 0.4 0.7 0.9 1.5 0.6 3.1
Nominal unit labour cost growth 3.0 2.3 -1.2 -3.1 2.3 -0.5
Real unit labour cost growth 0.6 -0.2 0.4 -1.5 0.0 -1.6

Sources: European Commission- Employment in Europe 2003 and EU Economy 2003 Review.

Pay

Pay is, of course, a central issue in collective bargaining in all three cases examined. According to EIRO estimates (TN0303102U), average collectively agreed pay increases across the current EU stood at 3.4% in 2002, down from 3.7% in 2001. This reversed an upward trend seen since 1999. There were considerable variations between Member States - from 5.4% in Greece to 2.1% in Austria. When the 3.4% increase is matched with average EU inflation of 2.1% over 2002, this indicates a real pay increase of 1.3% (down from 1.4% in 2001). Comprehensive figures are not available for the new Member States, but pay increases were generally above the EU 15 average.

In Japan, the average pay increase resulting from the annual spring bargaining round in 2002 was 1.66%, down from 2.01% in 2001 (2.06% in 2000 and 2.21% in 1999) and the lowest level since the relevant statistics were first published. Inflation stood at -0.9% in 2002, indicating a real collectively agreed pay increase of around 2.6% (unchanged from 2001). However, it should be noted that the spring wage increases, while having a major influence on general wage levels, are not directly reflected in the trend in average wages.

In the USA, collective bargaining plays a far less important role in setting overall national pay increases than in most EU countries or Japan, and indeed there is a significant differential between the pay of the one-sixth or so of workers represented by unions (and thus covered by bargaining) and that of non-union workers: the former's usual median weekly earnings were 26% higher than the latter's in 2002. Negotiated wage increases averaged 3.9% in 2002, unchanged from 2001. Adjusted for inflation, real negotiated pay increases amounted to 2.3% in 2002 (up from 1.1% in 2001).

Nominal collectively agreed pay increases were thus highest in the USA in both 2001 and 2002, followed by the EU and Japan. However, given its negative inflation, real agreed pay increases were highest in Japan in both years, with the EU in second place in 2001 and the USA in 2002.

Turning to changes in average earnings (a wider measure than collectively agreed pay increases), in the EU 15 (according to EIRO estimates), average earnings increased by an average of 3.9% in 2002 (an average of 12 countries), down from 4.3% in 2001. In the USA, the median usual weekly earnings of full-time wage and salary workers rose by 2.2% between 2001 and 2002, down from 3.6% over 2000-1. In Japan, total cash earnings fell by 2.4% from 2001 to 2002, following a fall of 1.1% in the previous year. Again correcting for inflation, it seems that real earnings in 2001 increased by most in the EU (1.8%), followed by the USA (0.6%) and Japan (-1.2%).

Looking at pay bargaining in 2002 more closely, in the EU 15 a trend towards increased pay moderation was relatively widespread, with the average rate of nominal increase falling in Austria, Belgium, Finland, Ireland, Luxembourg, the Netherlands, Portugal and the UK. However, the rate of increase rose in Germany, Greece, Italy and Spain, and remained stable in Denmark and Sweden. The rate of real pay increase rose in nine countries and fell in only five. In those countries where pay moderation continued or strengthened, this was largely as a result of sluggish economic performance, as well as the efforts of the government or social partners in some cases (eg the Netherlands, Belgium, Finland and Ireland). In some other countries, however, pay moderation faltered somewhat during the year - notably Germany. As well as increases, changes in pay structures (as in German metalworking) or greater pay flexibility (eg the Danish industry sector) were also on the agenda in some cases.

In Japan, given the poor economic conditions, the mood of the 2002 spring wage negotiations was that safeguarding jobs would take precedence over pay increases. For the first time, the Rengo trade union confederation broke from its tradition of presenting a unified demand for an increase in basic monthly pay and focused its demands on the two pillars of 'regular' annual pay increases and maintaining employment levels. Employers countered that cuts in total personnel expenses would be necessary to correct Japan's high cost structure and that it might, in some cases, even be necessary for employees to accept pay reductions. As well as the record-low average settlement in the pay round, this context also led to two distinct departures from previous bargaining rounds. First, discussions on 'work-sharing' led to a certain level of consensus being reached between unions, management and the government in March 2002 (see below under 'Working time'). Second, a number of leading electronics manufacturers, just a few days after agreeing to union demands for regular pay rises, reversed their position, proposing a postponement of the implementation of the increases and pay cuts within a set time-frame. A degree of agreement was reached between labour and management on these issues. Union acceptance of these proposals is believed to reflect the fact that employment anxieties are at an all-time high in this key sector.

In the USA, collectively agreed pay increases held up in 2002, compared with 2001, as the economy moved unevenly towards improved growth. Increases were particularly high in the strongly-recovering construction sector, with first-year contracts providing for a 4.5% average rise. However, the issue of health insurance arguably replaced pay at the top of the bargaining agenda (see the box below).

One feature of pay which is common to the EU, Japan and USA is the gender wage differential. In the current EU (plus Norway), women's earnings (based on hourly figures) were on average 80.8% of men's in 2002 (according to EIRO estimates - TN0303102U). In the USA, the equivalent figure for 2002 (based on the median usual weekly earnings) was somewhat lower, at 76.5%. In Japan, the figure ranged in 2000 (the latest year for which figures are available) from 64.3% for workers who are graduates of senior high schools to 75.5% for graduates of higher professional schools or junior colleges. Although the gender pay gap appears to be narrowing somewhat in Japan, the USA and most EU Member States, it remains persistent and relatively high.

Table 5. Pay developments in EU 15, Japan and the USA, 2001-2 (%)
. EU 15 Japan USA
. 2001 2002 2001 2002 2001 2002
Average nominal collectively agreed pay increase 3.7 3.4 2.0 1.7 3.9 3.9
Average real collectively agreed pay increase 1.4 1.3 2.6 2.6 1.1 2.3
Average nominal earnings increase 4.3 3.9 -1.1 -2.4 3.6 2.2
Average real earnings increase 2.0 1.8 -0.5 -1.2 0.8 0.6

Sources: EIRO, JILPT, Bureau of National Affairs and BLS. See main text for definitions.

Health insurance bargaining in the USA

Health insurance was probably the USA's most contentious bargaining issue in 2002, against a background of spiralling costs and a sharp jump in the number of people without such insurance. For working age people and their children, adequate health coverage is largely contingent upon employment, and unionised workers are up to 28% more likely to have employer-provided health insurance than those not covered by a collective agreement. Some 38% of all companies surveyed in 2001-2 said that they offered no health coverage (they are not legally bound to do do). These companies tend to be those with fewer than 200 workers. The number of uninsured people stood at 43.6 million in 2002 (15.2% of the population), up from 41.2 million (14.6%) in 2001 - the largest annual increase in a decade. In 2002, almost half the uninsured were working full time.

The increase in health costs and the competitive advantage enjoyed by companies not offering coverage to their employees motivated numerous employers to seek 'cost-shifting' and health coverage reductions at the bargaining table in 2002. Research published in September 2002 found the largest annual increase in premiums for job-based coverage since 1990, along with increases in deductibles and cuts in benefits. Premiums rose by 14% in 2002, the third successive year of double-digit increases. In 2002, 42% of employers went to the bargaining table hoping to gain larger contributions from employees for health insurance co-payments, deductibles or premium contributions, according to a Bureau of National Affairs (BNA) survey. Employers were generally successful in shifting some of the increases in healthcare costs onto their employees. For family coverage, the employee's share averaged USD 2,084 in 2002, up USD 283 from 2001. However, costs are rising so steeply that employer efforts to shift the burden, even if successful, are unlikely to relieve the pressure on companies for long.

Retiree health benefits have been particularly vulnerable. In 1988, some 66% of employers provided health benefits for their retired employees. In 2001, this figure had fallen to 34%, with a 9% drop in 2001 and 2002 alone. An announcement in February 2003 by Bethlehem Steel Corporation that it would terminate healthcare and life insurance benefits for 95,000 retired workers and their dependents because it could no longer afford to pay the coverage gave visibility to the growing crisis. Moreover, a decision by General Electric to shift some of its rising health coverage costs onto its trade unions - five months before collective bargaining was scheduled to begin - triggered a two-day strike in January 2003. There are signs that healthcare reform might again become an issue in the presidential elections of 2004. A nascent political alliance between companies and unions is already visible in outline, as both sides of the bargaining table come to believe that the problem of spiraling costs needs to be addressed politically. This may eventually result in unions and employers lining up against the private insurance companies, the pharmaceutical industry and the large private hospitals.

Working time

According to EIRO estimates (TN0303103U), the average normal weekly working hours set by collective bargaining across the EU (plus Norway) stood at 38.2 hours in 2002, unchanged from the 2001 figure. Agreed normal weekly hours ranged from 40 hours in Greece to 35.7 hours in France, but almost all other Member States were in the 37-39 hours range. In 2002, as in the previous four years, major working time reductions were absent across the EU and Norway, with the notable exception of France, where bargaining prompted by legislation has cut the normal week over recent years from 39 hours to 36.1 in 2001 to 35.7 in 2002, and, to a lesser extent, Belgium, where the 2001-2 national intersectoral agreement provided for a one-hour cut in the working week, from 39 to 38 hours, from January 2003. Even in France, the new conservative government which came to office in 2002 has sought to make the impact of the 35-hour week more flexible by increasing the annual overtime quota (FR0209105F). However, working time remains a key bargaining issue. Cuts were achieved in 2002 in some countries in particular sectors and companies (as in Greece, Italy, Luxembourg and Spain). Smaller reductions and shorter working time in the form of additional leave, often linked to the introduction of greater flexibility, were still common (as in Belgium and Denmark). Minor overall average reductions were seen in Spain, the UK and eastern Germany. Furthermore, working time reductions remain on the agenda of many trade unions in the EU, especially in those countries with above-average agreed weekly hours, such as Greece, Portugal and Spain.

In most EU countries, relatively high collective bargaining coverage levels mean that the average collectively agreed normal working week is probably relatively close to the actual normal working week (though this is less likely to be the case in a country such as the UK).

Comprehensive information is not available on average collectively agreed working time in the new Member States. However, agreed weekly working time generally appears to be above the average for the EU. Working time reductions seem high on the industrial relations agenda in some acceding countries, such as Hungary and Slovakia.

Figures on collectively agreed normal weekly hours are not available from Japan and the USA, but some comparisons can be made with the former. For Japan, the figure for 'average scheduled weekly working hours'- determined by collective agreements or 'work rules' (uniform rules on working conditions that employers are required by law to compile) - was 39.4 hours for full-time workers in 2002 (1.2 hours higher than the EU 15 average figure), a 0.2-hour increase on 2001. However, scheduled working hours are decreasing on a long-term basis. Working hours are a major theme for company-level bargaining in Japan, while working time reduction was on the agenda of the national social partners in 2002, with a degree of tripartite consensus reached on maintaining employment by sharing out employment opportunities though cutting working hours ('work-sharing'). It was agreed to continue to promote diversified employment patterns such as part-time work, while establishing work environments that would accommodate such workers. The accord also included five basic principles for the introduction of work-sharing, incorporating measures such as 'emergency' work-sharing, aimed at tackling the current poor employment situation.

No relevant data are available on collectively agreed working time in the USA, where the issue currently seems to have a relatively low profile in collective bargaining.

It should be noted that all EU Member States have a framework of legal rules on maximum working times, dividing fairly equally between those that set the statutory maximum working week at 48 hours and those which set it at 40 hours. In Japan, the normal working week is set by law at 40 hours though, as is frequently the case, extension of working hours may be specified in individual labour-management agreements, although guidelines issued by the Ministry of Health, Labour and Welfare set a maximum limit. In the USA, there is no statutory maximum or normal working week as such, though employees covered by the Fair Labor Standards Act must receive overtime pay for hours worked in excess of 40 in a working week (at a rate not less than 50% over their regular rates of pay).

Comparison of actual hours worked is very difficult, owing to a lack of uniform statistics. According to Eurostat's 2002 Labour Force Survey, the average usual weekly hours worked by full-timers in the current EU stood at 40.0 (varying between 43.3 hours in the UK and 37.7 in France), compared with 40.1 in 2001 and 40.3 in 2000. The same survey found that the average usual weekly hours worked by full-timers in the 10 new Member States stood at 41.4 (ranging between 43.6 hours in Latvia and 39.5 in Lithuania) In the USA, full-time workers - defined as those working at least 35 hours per week - worked an average of 42.8 hours per week in 2001 (the latest, year for which figures are available), compared with 43.3 hours per week in 2000. The Japanese figures for actual hours worked do not distinguish between full- and part-timers and are expressed in monthly, rather than weekly, terms. However, the figures indicate a slight decrease between 2001 and 2002.

Perhaps the best way of comparing working time between the EU, Japan and USA, and of assessing trends, is to look at annual working hours. These are influenced both by the level of weekly hours and the amount of paid annual leave and public holidays.

Paid annual leave is considerably higher on average in the current EU (no figures are available for the new Member States) than in Japan or the USA. In 2002, the average number of days of collective agreed annual leave in the EU plus Norway (harmonised on the basis of a five-day working week) stood at 25.9, up from 25.7 days in 2001 (according to EIRO estimates). However, it should be noted that agreed annual leave entitlement varies considerably within the EU, from 31.3 days in the Netherlands to 20 days in Ireland. For Japan, figures are available from JILPT on the average paid holiday entitlement, which stood at 18.1 days in 2002, up from 18.1 days in 1999. However, a notable feature of holidays in Japan (which is not known to be a major issue in the EU), is that by no means all of this leave entitlement is actually taken up. In 2002, only 48.4% of annual leave entitlement (8.8 days) was taken on average, down from 49.5% (or 8.9 days) in 2001. This rate of take-up has been decreasing as leave entitlement has increased over the past decade. The latest figures available for the USA from BLS refer to 1997 and cover 'average vacation days' in the medium and large private sector. No overall figure is available for all workers, but separate ones for varying lengths of service. Thus, annual vacation days stood at 9.6 days at one year's service, 13.8 days at five years' service, 16.9 days at 10 years' service and 20.3 days at 20 years' service. All these figures had increased slightly over the previous decade.

All current EU Member States have a statutory minimum period of paid annual leave of between 20 and 25-day minimum entitlement, with an average entitlement of just over 22 days (harmonised for a five-day week). In Japan, legislation entitles workers with at least six months' continuous service to a minimum of 10 days' paid annual leave, with the number of days increasing with the length of service. There is no statutory annual leave entitlement in the USA.

With regard to public holidays, on average current EU countries have just under 11 per year at present (varying between eight in the Netherlands and 13 in Austria, Portugal and Spain). This compares with 14-15 in Japan and 10 in the USA.

Broadly comparable figures are available for the current EU (plus Norway) and Japan on agreed/scheduled weekly hours, annual leave and public holidays. On this basis, average normal annual hours, before taking into account leave and holidays (and assuming a 52-week year) stood at 1,986.4 in the EU in 2002 and 2,048.8 in Japan. Reducing these figures by the total average paid leave entitlement and public holidays (36.1 days in the EU and 33.1 in Japan) produces average normal agreed/scheduled annual hours of 1,710.6 in the EU and 1,788.0 in Japan. Insufficient data are available to produce a similar figure for the USA. However, JILPT has produced a figure for 'scheduled' annual working hours in the USA in 2000, which stood at 1,747. This suggests that normal annual hours are highest in Japan, followed by the USA and the EU. However, individual current EU Member States - Greece, Ireland and Belgium - have normal annual hours that are higher than Japan's.

With regard to hours actually worked per year, comparisons are again difficult. However, based on various national figures, JILPT has produced estimates for annual total hours actually worked (by production workers in manufacturing industry) for Japan, the USA and three of the largest EU countries - France, Germany and the UK. Thus in 2000, actual annual hours worked stood at 1,986 in the USA, 1,970 in Japan, 1,902 in the UK, 1,589 in France and 1,525 in Germany (1999) figure.

Legislative developments

Arguably a key point of industrial relations difference between the EU and many of its Member States on one hand, and Japan and the USA on the other, is the nature and extent of employment law. At both EU level and in many individual Member States, there is detailed legislation regulating many areas of industrial relations, employment conditions and workers' rights, which is added to frequently as situations and priorities change. In Japan and particularly the USA, by contrast, only the basic 'rules of the game' and/or a number of minimum rules on employment conditions are laid down in law.

Thus in 2002, the EU adopted Directives on: minimum requirements for the right to information and consultation of employees in undertakings and establishments (EU0204207F); the working time of persons performing mobile road transport activities (EU0204208F); the protection of employees in the event of the insolvency of their employer (EU0112245F); equal treatment for men and women as regards access to employment, vocational training and promotion, and working conditions (EU0205201N); and the minimum health and safety requirements regarding the exposure of workers to the risks arising from physical agents (noise) (EU0212202N). The European Commission also proposed a new Directive on the working conditions of temporary (agency) workers (EU0204205F) and consulted the social partners on possible Community initiatives on anticipating and managing change (EU0201235F) and the protection of workers' personal data (EU0211206F).

At national level, the usual plethora of new employment-related legislation was proposed or adopted in the current Member States in 2002 (in some cases implementing EU Directives) (TN0303101F) covering matters such as: part-time work, fixed-term contracts and temporary agency work (eg in Denmark, Germany, the Netherlands and Sweden); equality, non-discrimination, and reconciliation of work and family life (eg in Austria, Belgium, Denmark and the UK); working time (eg in France and Luxembourg); employment, labour market and job creation (eg in Denmark, France, Germany, Italy and Portugal); health and safety (eg in Finland and Portugal); social security (eg in Denmark, Portugal, Spain and Sweden); termination of contract (eg in Austria and France); and general industrial relations issues (eg in Portugal and the UK). In many new Member States, recent years have seen hectic employment law activity as the countries harmonise their provisions with EU norms. For example, 2002 saw major new legislation in countries such as Hungary, Poland and Slovakia.

By contrast, in the USA, there appears to have been an absence of significant and specific new legislation on employment and labour issues at federal level in recent years. In 2002, the main legislative developments with an impact on industrial relations were essentially aimed at other issues - the 'war on terrorism' and privatisation. Notably, the creation of a new Homeland Security Department involved the introduction of new policies on recruitment, pay, collective bargaining, performance appraisal, discipline and dispute resolution for the new agency's 170,000 employees (drawn from 22 pre-existing government departments), which trade unions feared would reduce their collective bargaining rights. The new law creating the Department instructed its leading officials to consult with employee organisations before making changes to civil service rules. One of the other most notable legal developments was an unprecedented government intervention, invoking the 1947 Taft-Hartley Act, to end a lock-out in Pacific coast ports as a threat to national security.

Japan has seen rather more new employment legislation in recent years (driven by various socio-economic changes), but its volume has arguably been considerably less than in the EU. In 1999, for example, several important provisions of the Labour Standards Law were amended, covering matters such as fixed-term employment and the content of employment contracts, while the law on private employment agencies and temporary work was largely deregulated. Furthermore, gender equality legislation was strengthened. Revised legislation on the employment of older workers was introduced in 2000. The next year saw new or amended legislation on: the continuation of employment contracts following the division of a company into separate organisations; employers' obligations in collective redundancies (see below under 'Company restructuring'); and the resolution of individual labour disputes. However 2002 saw virtually no significant legislative developments, except for a proposal for amendments to the Labour Standards Law, related to dismissal (see below under 'Company restructuring').

Company restructuring

A common trend in recent years across the EU, Japan and USA has been towards company restructuring. This can take many forms, including: mergers and acquisitions; splits or demergers; divestments; bankruptcy or total closure; closure or scaling back of sites and operations; national or international relocation or outsourcing of operations; and internal restructuring. The increasing amount of such restructuring is often attributed to underlying processes of adaptation to economic globalisation and the internationalisation of competition and markets, as well as to changing market and economic conditions. For the industrial relations arena, the key issue in restructuring is essentially its effect on jobs - both their quantity and their quality. However, there are considerable variations between countries in terms of: the extent to which restructuring is a prominent theme in industrial relations; the regulatory framework for dealing with the employment and social effects of restructuring; and the role of employee involvement and collective bargaining in this area.

Trends and perceptions

Restructuring has without doubt been one of the most high-profile topics in industrial relations in the current EU in recent years. The Renault Vilvoorde affair of 1997 (when the French-based motor manufacturer closed down its Belgian plant without adequate prior employee information and consultation, with the loss of over 3,000 jobs - EU9703108F) marked the beginning of a period in which the issue has received arguably unprecedented attention. Since then, each year has seen its quota of prominent cases of company reorganisation, with the most controversial being those which have involved major job losses, especially where prior information and consultation of the employees, and/or the provision of accompanying measures (retraining, outplacement, financial arrangements etc) are perceived as inadequate. Cases of multinational companies relocating operations and jobs to lower labour-cost countries (within the EU, in central and eastern Europe or elsewhere in the world) have been especially contested. In a number of cases, job losses following mergers and acquisitions have been widely criticised. Furthermore, there has been a steady stream of internal restructuring exercises and accompanying workforce reductions to deal with changing market conditions, which has intensified in the past few years as economic conditions have worsened, particularly affecting sectors such as telecommunications, information and communications technology (ICT), civil aviation and motor manufacturing.

It should, however, be noted that the extent and nature of restructuring varies significantly across EU countries. This has been linked by a 2002 EIRO study (TN0209101S) to varying national systems of 'corporate governance' (ie the set of mechanisms that control and influence senior management). In one group of EU countries (essentially Ireland and the UK), shareholdings are dispersed across a range of financial intermediaries and there is a well-developed 'market for corporate control'- the 'outsider' system of corporate governance. In most continental European countries, there is more stability in ownership and less of a market for corporate control - the 'insider' systems (though these vary considerably). However, many countries that have traditionally been characterised by insider systems are evolving (though none has yet converged closely towards the UK-Irish model). Countries such as the Netherlands, Finland, Spain and France have witnessed significant shifts in the direction of a more outsider-oriented system, while in others such as Germany and Sweden the changes have been notable. The study concluded that restructuring has been most pervasive, and has occurred most rapidly, in the outsider systems and in rapidly evolving insider systems.

The increased focus on restructuring since the late 1990s has led to further EU-level initiatives (as well as national ones in countries such as France - FR0311106F). For example, the European Commission set up a high-level group on the economic and social implications of industrial change (the 'Gyllenhammar group'), which reported in 1998. It recommended that: companies should assume the main responsibility for anticipating change; companies with large numbers of employees should publish reports on managing change; and the Commission should support this process by setting up an 'observatory' on change. This latter point was followed up with the creation in 2001 of the European Monitoring Centre on Change (EMCC), run by the European Foundation for the Improvement of Living and Working Conditions, which seeks to enable the exchange of practice, information and ideas on the management and anticipation of change. Furthermore, in January 2002, the Commission launched consultations with the EU-level social partners on establishing EU-level principles to underpin 'socially intelligent' corporate restructuring. This process led in October 2003 to the conclusion by the social partners of a joint text entitled 'Orientations for reference in managing change and its social consequences'.

In Japan, restructuring and the 'employment adjustments' which often accompany it have become an increasingly important issue in recent years, following a long period of economic growth during which employment stability and 'lifetime employment' were the norm, at least for 'regular' workers in larger companies. Traditionally, when Japanese companies (although this does not apply to all businesses) have an excessive number of staff, they do not rush into personnel cuts, but instead adopt measures such as reducing non-scheduled working hours, not renewing the fixed-term contracts of non-regular employees, and limiting new recruitment, whilst monitoring their business environment. Only if, having taken these steps, there is no evidence of a turnaround will they then start to shed regular workers. Even when dismissals become inevitable, employers do not usually name specific workers to lose their jobs, with large companies instead tending to seek applicants for voluntary and early retirement, offering them a 'golden handshake' in the form of an extra pension allowance, for example. There is a widespread view that companies will start looking to cut their workforces after two consecutive years of losses

However, since the late 1990s, in the face of continuing poor economic conditions, the number of Japanese companies engaged in restructuring and workforce 'rationalisation' has been increasing and even those companies which in the past strongly supported employment stability for regular workers have begun to show signs of a change of direction. This tendency strengthened in late 2001 when, with the global economy facing difficulties, a succession of plans for personnel reductions were announced. The scale of the workforce-reduction plans announced in the recession-hit ICT sector was particularly remarkable, with some tens of thousands of workers slated for redundancy. However, it should be noted that many of the job losses were to be effected in multinationals' overseas establishments (as at Hitachi and Fujitsu) and, even when they were effected domestically, there were many cases of opting for non-replacement of employees retiring normally, so the proportion of workforce reduction through means such as voluntary retirement schemes was not so great. For example, Toshiba announced plans to spin off its semiconductor and memory-chip business, which would merge with another company and in the process shed some 17,000 workers in Japan. Approximately 10,000 of these jobs would be lost due to 'natural wastage', with an early retirement scheme to account for the remaining 7,000 positions. Such workforce reduction plans were less frequent in 2002, but many companies continued to reduce their workforces as a means of countering the effects of long-term economic malaise. A survey by the Ministry of Health, Labour and Welfare found that in the three-year period 2000-2, around 8% of companies 'solicited applicants for voluntary and early retirement', with the percentage increasing each year.

The USA has also faced major corporate restructuring over recent years. However, the issue does not yet appear to have achieved the same prominence or caused the same debate as in the EU and Japan. It has been argued that constant restructuring is widely seen as normal in the USA, with a perception in business and policy circles that it is almost always good for the economy and competitiveness (if difficult for some workers in the short term). Although developments such as plant closures have caused controversy and opposition in the past, the issue has not generally had a high profile. However, there are some indications that recent trends towards relocation of operations outside the USA and 'global outsourcing', with attendant job losses, may be pushing the restructuring issue somewhat further up the political agenda.

Regulation

The EU has long sought to regulate a number of employment and social aspects of restructuring, adopting Directives on such issues as long ago as the 1970s. There are Directives in place (revised over the years in some cases) covering matters such as collective redundancies, business transfers and insolvencies. These do not attempt to prevent or control such actions (though mergers, for example, are in some cases subject to EU-level approval on competition grounds), but to lay down certain rules on employers' obligations and employees' rights (including information and consultation). For example, the collective redundancies Directive provides that employers contemplating collective redundancies must begin consultations with worker representatives in good time with a view to reaching an agreement. These consultations must at least cover ways and means of avoiding collective redundancies or reducing the number of workers affected, and of mitigating the consequences by recourse to accompanying social measures aimed, for example, at aid for redeploying or retraining workers made redundant. To enable worker representatives to make constructive proposals, the employer must in good time during the consultations supply them with all relevant information. Employers must also notify the competent public authorities in writing of any projected redundancies. In addition to these specific provisions, there are Directives providing for employee information and consultation rights on company restructuring-related matters at both European and national level. These various items of EU legislation have been implemented in the current EU Member States (and must be transposed by the new Member States by the time of accession), and there are various additional provisions in a number of countries (see below). In a few Member States, for example, administrative or judicial authorisation is required for employers to make dismissals (notably the Netherlands - NL0308104F).

In Japan, there has historically - doubtless reflecting its long period of economic growth and employment stability - been an absence of specific legislation on redundancies and other outcomes of restructuring, or on employee information and consultation in such circumstances. No formal regulations existed in labour law to restrict employer-instigated dismissals, though the accumulation of legal precedents led to the establishment of the legal principle that dismissal without rational cause is contestable. The impact of this legal principle and the high priority assigned to maintaining and preserving employment levels in companies have long meant that dismissals are generally handled with caution. However, in recent years, the protracted economic malaise has forced many employers to resort to direct 'labour shedding'. Against this background, there was discussion of incorporating regulations on dismissal into the Labour Standards Law. In December 2002, the Labour Policy Council issued a number of recommendations, including the view that: 'Case law establishing a doctrine of abuse of the right of dismissal must be written into the Labour Standards Law and a regulation to the effect that if a company dismisses an employee without a specific objective and a rational reason, and if such dismissal is recognised to be inappropriate in light of regulations such as the Labour Standards Law, the company shall be regarded as having abused its rights, and such dismissal shall be invalidated, incorporated therein.' In response to these proposals, the government presented a draft bill incorporating the gist of the Council's recommendations, and the relevant legislation was adopted in 2003, regulating dismissals for the first time.

The new dismissals provisions are not the only restructuring-related legislation introduced in Japan in recent years. In 2001, the Employment Measures Law and other laws were amended with the aim of promoting smooth re-employment in response to changing socio-economic conditions. A notable change was that employers which make an employment adjustment that generates a large number of redundancies at one time are required to adopt a plan to support re-employment systematically ahead of termination. Also in 2001 a new law laid down rules concerning the continuation of employment contracts in the event of the division of a company into separate organisations. The law stipulates that, in cases where the contracts of workers affected by the division are not carried over to the new organisation, those workers can lodge an objection and have the employment contract continued.

In contrast to the EU, and now Japan to some extent, US companies can generally restructure and dismiss workers with virtually total freedom. The only significant legislation regulating any aspect of these issues is the Worker Adjustment and Retaining Notification Act (WARN Act), which requires companies with 100 or more full-time employees to provide 60 days' notice in advance of a plant closure or mass redundancies. The WARN Act also provides for notice to be given to States' 'dislocated worker units' so that dislocated worker assistance can be promptly provided. There are also some laws, such as the Trade Readjustment Act, that provide some relief and assistance to certain categories of 'displaced' worker.

The involvement of employees and collective bargaining

As indicated above, an important feature of corporate restructuring in the EU is that, in all current Member States, certain aspects of it are subject to the statutory involvement of employee representatives (including trade unions in some cases), as a result of both the implementation of relevant EU Directives and domestically generated law (TN0107201S and TN0102401S). The aspects of restructuring subject to employee involvement include collective redundancies and business transfers in all cases, plus in most countries a variety of workplace-related restructuring issues (such as changes in work organisation, changes with consequences for employment structure/levels, or the introduction of new working or production methods) and wider company restructuring issues (such as closures and cutbacks in operations and activities, mergers and/or acquisitions and/or divestments, and changes in company structure).

As highlighted by the recent EIRO study on restructuring and corporate governance mentioned above, there are two principal channels through which employees in the EU can influence restructuring. The first is by using rights based in systems of co-determination or 'social concertation'. Some national systems of employee representation grant employees the right to be informed well in advance of any restructuring that will have a significant impact on employment, and allow their representatives to negotiate a 'social plan' to deal with the consequences. This is the case in Austria, Belgium and Germany. Many countries in Europe have rights concerning employee representation on company's supervisory or management boards, or boards of directors (TN9809201S), which have the ability to discuss and decide on proposed cases of restructuring. This is the case in Sweden, Austria, Germany, the Netherlands, Denmark, Finland and Luxembourg. To varying degrees, and in various forms, these countries have systems of co-determination that are part of a tradition of dialogue and that promote a spirit of compromise when it comes to the extent and nature of restructuring. Employee representatives are commonly drawn from trade unions, which can also be influential through collective bargaining in these systems, of course. Indeed, the position of unions is in most cases strengthened by the institutions of co-determination.

The second type of EU system is where the source of employee influence over restructuring is primarily through trade unions and collective bargaining. In some countries, this channel is the primary way in which employees are able to exert pressure on management, despite the existence of works councils. This is the case in France, Italy, Spain, Portugal and Greece. In the remaining, two countries, the UK and Ireland, which currently lack formal institutions such as works councils and do not have a strong tradition of social partnership, the influence of employees is largely dependent on the strength of unions at firm level. These two countries at present have a fairly minimalist legal framework, with the legal rights that employees enjoy coming mainly from the EU Directives on collective redundancies and business transfers. In countries where employee influence is principally through unions and collective bargaining, management-employee relations on restructuring issues tend to be more adversarial, and the ability of employees to influence restructuring varies considerably from sector to sector and firm to firm according to union strength.

At European level, some 700 multinational companies are thought to have EWCs in place, and company restructuring is often a key issue in the formal remit of these information and consultation bodies. While commentators have criticised many EWCs for playing a weak or symbolic role - and alleged failure by management properly to inform and consult EWCs on restructuring has been a key issue in a number of controversial cases - there is evidence of genuine dialogue in some cases, and even the conclusion of Europe-wide agreements on restructuring issues in a few 'best-practice' EWCs (such as those at Danone and General Motors).

Thus, with considerable national differences, aspects of corporate restructuring are subject to employee involvement (often based on statutory provisions) and often company/local-level collective bargaining in the current EU. The importance of some form of employee involvement in this area seems to be widely accepted by all main actors in the EU, including employer organisations. For example, the October 2003 European social partners' joint text on restructuring underlines that 'the existence of a good social dialogue in a climate of confidence and a positive attitude to change are important factors to prevent or limit the negative social consequences' of change. It also highlights the need for 'good information and consultation of the workers and/or their representatives throughout the process of change'.

Turning to Japan, restructuring is a key issue in company-level collective bargaining and consultation. However, there is no legislation which requires information, consultation or negotiations on restructuring, redundancies etc. A survey published by the Ministry of Health, Labour and Welfare in June 2003 found that, of the discussions undertaken by trade unions and management during the previous three years, those on 'corporate restructuring and/or reductions in business divisions' had accounted for 42.3% of the total - ie a 3.4 percentage point increase on the previous survey (1997). In terms of the stage at which negotiations are initiated, on the employer side 41.6% of negotiations are commenced once 'the broad framework has been decided', followed by 36.4% at the 'exploratory phase' and 13.5% once 'the details have been fixed'. Thus employers are tending to initiate negotiations with labour representatives at a comparatively early stage in the restructuring process. A survey of collective agreements by the same Ministry (covering 5,000 trade unions), released in June 2001 for the first time in five years, found that advance discussions relating to the reduction and/or closure of a business had increased, and these matters appeared in 40% of all agreements reached. Furthermore, a 2000 survey by the Ministry of Health, Labour and Welfare found that issues such as 'temporary lay-off, workforce rationalisation and dismissal' are dealt with by a majority of the labour-management consultative bodies which exist in around 40% of private-sector establishments with more than 30 employees (see above under 'Employee invovlement').

Above company level, restructuring-related issues have also been discussed by the national-level Japanese social partners over the past couple of years. With the desire to avert job losses and safeguard jobs apparently weakening among Japanese companies, the Rengo trade union confederation and the main employer confederation engaged in vigorous discussions on the importance of measures to maintain and preserve employment levels in 2002. In March, the two organisations reached a broad agreement to stabilise employment by focusing on measures such as 'work-sharing' (see above under 'Working time').

In the USA, there is no legislation requiring employers to inform, consult, or negotiate with employees over restructuring, with the exception of the stipulation in the WARN Act that larger companies must give 60 days' notice of plant closures or mass redundancies. The general deregulatory climate, coupled with the widespread positive perception of restructuring (see above) means that changes to this general picture seem unlikely at present. In terms of collective bargaining, a handful of collective agreements contain agreed procedures related to company restructuring, relocation and closure, but these tend to be the exception to the general rule. Rather more widespread are collectively agreed provisions on some aspects of job security. Research from the Bureau of National Affairs (BNA) indicated that around 95% of current collective agreements in early 2001 contained provisions designed to enhance workers' employment security - such as extended 'recall rights', subcontracting restrictions, or advance notice of shutdowns. In terms of other types of employee involvement, there is no legislation in the USA on employee representation through works council-type bodies or any comparable arrangements (see above under 'Employee involvement'). Where employee participation arrangements exist on a voluntary basis, they tend to be relatively restricted in scope.

New forms of work

A growth in 'atypical', 'non-regular' or 'new' forms of employment - such as temporary work (mainly fixed-term contracts and temporary agency work), part-time work, teleworking, 'economically dependent' work or on-call work - is a common long-term trend across the EU, Japan and the USA (though definitions vary). This has been accompanied by a general decline in 'traditional' full-time employment on open-ended contracts (at the employer's premises). However, all new forms are not growing equally in all countries, while the USA especially has seen some fall-off in the overall trend towards 'atypicality' (which tends to retreat in times of high employment) and the phenomenon is generally less developed in the new EU Member States. According to labour force survey statistics, between 1997 and 2001, the percentage of 'regular' employees in the overall Japanese workforce declined from 76.8% to 71.3%, while in the USA the proportion rose from 70.6% in 1995 to 73.4% in 2001. Table 6 below gives recent comparative figures for part-time and temporary work.

Table 6. Extent of part-time and temporary work in Europe, Japan and USA (% of employment)
. Part-time work Temporary work
EU 15 current Member States 18.2% 13.1%
EU 10 new Member States 7.8% 11.1%
Japan 14.3% 14.4%
USA 17.5% 4.9%

Notes: EU - figures from 2002 Eurostat Labour Force Survey, part-time work self-reported, temporary work refers to contracts of limited duration; Japan - figures for 2001 from JILPT, part-time work is less than 35 hours per week, temporary work figure is sum of casual workers, temporary agency staff and contract workers; USA - figures from BLS (December 2002 for part-time work, February 2001 for temporary work), part-time work is less than 35 hours per week, temporary work refers to 'contingent' work ('workers who do not expect their jobs to last') plus temporary agency workers.

Part-time work (despite differing definitions) is at around the same level in the current EU and the USA, at something over one-sixth of all employment (though the EU figure masks huge variations between countries, with the rate as high as 43.8% in the Netherlands and as low as 4.5% in Greece), and a few percentage points lower in Japan. The level in the new EU Member States is considerably lower, at under half the current EU figure. A common factor is that part-time work is predominantly a female phenomenon - 33.5% of female employees are part-time in the EU (2002 Eurostat figure), some 43% in Japan (2001 JILPT figure) and around 25% in the USA (2002, BLS). At 10.3% in 2002, the proportion in the new EU Member States is considerably lower. Some 79.3% of all part-timers are women in the current EU and 60.3% in the new Member States (2002, Eurostat) compared with 68% in the USA (2002, BLS) and 70% in Japan (2000, JILPT).

Temporary work of various kinds is much more common in the current EU and especially Japan than in the USA, with the new EU Member States in an intermediate position (though problems of definition are particularly acute in this form of employment).

Teleworking/telecommuting is on the increase in all three cases, though again differences in definition may make comparisons difficult. According to the European Commission, there are about 4.5 million employed teleworkers in the present EU (around 2.8% of all employment) and about 10 million teleworkers in total. Telework is not yet, however, thought to have developed significantly in the new Member States. Japanese research (from Japan Telework Association) finds that the estimated number of employed teleworkers increased from 2.46 million in 2000 to 3.11 million in 2002 (5.7% of total employment), while self-employed teleworkers stood at 0.97 million in 2002 (8.2% of total self-employment). It is predicted that the number of employed teleworkers in Japan will nearly double to 5.63 million by 2007 (the survey defines teleworkers as those who are engaged in paid work on a regular basis, utilise ICT at work, work either in more than one workplace where they utilise ICT, or in only one workplace that is other than their employer's premises; and work outside their employer's premises for more than eight hours per week). In the USA, the number of employed teleworkers was put at 23.5 million, or around a sixth of all employment, in 2003 (plus 23.4 million self-employed teleworkers) by the International Telework Association and Council. The number of employed teleworkers had stood at 11.6 million in 1997 (these statistics are based on the wide definition of employees working at home during business hours at least one day per month).

Other new forms of employment are particularly relevant in some countries. For example, 2 million US workers (1.5% of total employment) work on-call - ie they are called to work only as needed.

There are a number of common reasons often cited for the overall growth of these 'new' forms of work, such as employers' wish to achieve flexibility and employees' wishes to balance their work and family/private responsibilities or to gain entry to the labour market. The situation varies between countries and forms of work in terms of the extent to which employees positively choose new forms of work or accept them because they cannot find 'regular' work.

Regulation and industrial relations

The issue of new forms of work arguably highlights clearly the differences between the industrial relations systems of the EU, Japan and the USA. The EU (with national variations, of course) tends to regulate such forms of work specifically through legislation and, in some cases, collective bargaining, while the USA leaves the matter virtually unregulated, and Japan stands somewhere between the two positions.

A framework regulating a number of new forms of work is being created at EU level, seeking to prevent discrimination against the workers involved and promote quality and flexibility in such work. Directives, based on social partner agreements, have regulated part-time work (1997) and fixed-term work (1999), while a draft Directive on terms and conditions for temporary agency workers (EU0204205F) is currently under discussion (though negotiations are proving difficult - EU0306206F). Furthermore, in July 2002 the European-level social partners concluded an agreement on teleworking which is being implemented by the national member organisations of the signatory parties. The overall thrust of EU regulation is to achieve equal treatment between 'atypical' and 'typical' workers, as far as this is possible.

At national level, all current Member States regulate various aspects of new forms of work, with implementation of the abovementioned EU Directives leading to legislation even in countries which formerly had no regulation in this area. While there has been a widespread tendency to promote and permit various new forms of work (eg in some countries, such as Greece and Italy, temporary agency work has been formally allowed only in recent years), this has generally been accompanied by the establishment of a number of rules for their operation. Changes have occurred in some Member States with a high level of regulation in areas such as fixed-term contracts (eg relaxing the grounds on which they may be concluded, or the rules on their duration), but this has rarely been total deregulation, but rather a more flexible regulation.

In the new EU Member States, the regulation of various new forms of work has been subject to rapid change in many cases in preparation for EU accession, including the implementation of the relevant EU Directives. This has involved both new regulation and relaxation of previous restrictions. For example, new legislation has been adopted in the past couple of years on part-time and fixed-term work in Hungary (HU0308101F), and temporary agency work (PL0308103N) and fixed-term contracts (PL0311108F) in Poland.

Collective bargaining also plays a role in regulating new forms of work in most countries of the current EU. As seen above, the issue has been an important one in negotiations between the European-level social partners. At national level, collective agreements have varying roles in this area. Aspects of part-time work are covered by agreements at various levels in countries such as Austria, Belgium, Denmark, Germany, Italy, the Netherlands and Spain. These include: assisting conversion into full-time contracts; maximum and minimum working times and overtime; part-time work for older workers; rights to reduce working time; and training provisions. Some collective agreements (mainly sectoral) include provisions on fixed-term contracts in countries such as Denmark, France, Italy, the Netherlands, Spain and Sweden, dealing with matters such as the maximum proportion of such contracts, limits on their duration and rules on circumstances in which they may be used (TN0202101S).

While collectively agreed provisions on part-time and fixed-term work are often relatively well established, a more recent trend is towards collective bargaining relating to temporary agency work. Specific sectoral collective agreements for the temporary agency work sector have been concluded in recent years in Austria, Belgium, France, Luxembourg, the Netherlands, Portugal, Spain and Sweden. These agreements tend to cover issues such as pay and conditions, benefits, representation rights and training. There are also some company-level agreements between individual temporary work agencies and trade unions, as in Germany at Randstad Deutschland (DE0105222N) and Adecco. Another approach to temporary agency work is agreements in other sectors or companies which regulate the use made of agency work. This is the case, for example, in many Italian sectoral agreements, which: identify the cases when the use of temporary agency work is allowed or not allowed; and lay down the maximum number of agency workers permitted, as a proportion of the permanent workforce.

Teleworking is another issue which is increasingly subject to bargaining in some countries. Agreements - generally providing for equal treatment with other employees and regulating matters such as the voluntary nature of telework, training, equipment, monitoring and health and safety - have been signed, for example in: Austria (eg in electricity supply, telecommunications, information technology services, mineral oil production and information consulting); Denmark (eg in finance and commerce); Germany (eg in the rail sector and at Deutsche Telekom); Italy (eg in commerce and public administration and for SMEs); Sweden (eg in trade, commerce and services); and the UK (eg at British Telecom). The implementation of the EU-level social partners' agreement on telework seems to be stimulating further joint approaches on this theme in the Member States.

Overall, it can be said that new forms of work are without doubt a central issue in industrial relations in the EU.

In Japan, as seen above, the traditional model has been one of regular or lifetime employment, though some aspects of atypical work have been subject to legislative regulation - for example, there are limits on the duration of fixed-term contracts. However, there have been some changes as employment patterns have diversified. In 1993, a law was adopted on the 'improvement of employment management of part-time workers' (those who work shorter weekly scheduled hours than regular workers employed at the same establishments). It requires employers to promote effective utilisation of part-time workers' abilities, with due consideration of their actual working conditions and the balance with full-time workers, by taking effective measures in order to secure them proper working conditions and improve their management. The government is required to promote the human resource development and training of part-time workers, and to provide them with employment information and services. Guidelines were issued based on this law covering issues such as: changing the working schedules of part-timers; notice periods for part-timers with over a year's service; part-time work for older people; and preference for part-timers in filling full-time jobs. Employers should also seek to give equal treatment to part-time workers who work hours similar to full-time workers.

The operation of temporary work agencies was formerly tightly regulated (eg they could handle only a specified list of occupations), but in 1999 it was largely deregulated in response to pressures generated by structural change in the labour market and adoption of the relevant International Labour Organisation (ILO) Convention (No.181). At the same time, new legal provisions were introduced to ensure appropriate conditions of employment for temporary agency ('dispatched') workers. These included limiting the period of continuous employment for agency workers in the same job to a maximum of one year and requiring firms which have employed a dispatched worker on a continuous basis for one year, to make every effort to give preference to such a person when recruiting.

Telework is not subject to specific regulation in Japan, and teleworkers are in a somewhat ambiguous position in terms of whether they are covered by employment legislation, such as the law on minimum wages. It appears that teleworkers engaged in industrial production at home may enjoy legal protection, but not many of those who work in the service sector or information-related activities.

Despite the existence of some legislative provisions as outlined above, the topic of equal treatment of atypical workers has not yet been fully addressed in Japan. Surveys of part-time workers, for example, indicate that they report problems such as low wages and difficulties in taking paid leave. In many Japanese companies where job descriptions/demarcations are not always clear and regular employees are normally paid on a monthly basis whereas part-timers are paid on an hourly basis, the concept of 'equal pay for equal work' is not prevalent. However, the equal treatment issue is achieving a higher profile, at least with regard to part-timers, especially as they are increasingly occupying higher-responsiblity positions. Recently, a tendency has been observed to search around for an ideal form of equal (balanced) treatment of different types of workers, based on Japanese traditions. These issues have been discussed in a study group set up by the Ministry of Health, Labour and Welfare, which reported in July 2002. Trade unions are increasingly seeking to organise part-time workers and calling for their equal treatment. With regard to teleworking, both the Rengo trade union confederation and Nippon Keidanren employer organisation have been showing greater interest in the matter of late, with Rengo calling for the establishment of fair work rules for teleworkers, and Nippon Keidanren promoting diversity in workplaces.

The decentralised nature of collective bargaining in Japan makes it difficult to assess the extent to which issues related to new forms of work are covered by collective agreements. However, survey data do not indicate that such matters are often a specific issue for company-level bargaining or consultation.

The USA appears to have no specific legislation to regulate any new forms of employment, with employment law applying to all workers (though labour law restricts the right of some categories of 'contingent' workers to form unions and bargain) within a context of a generally low level of regulation of any employment matters. Equal treatment and opportunities legislation applies to issues such as race, sex, disability and age rather than to employment status. 'Contingent' or 'non-standard' workers tend to receive lower wages (especially part-timers) than those in traditional 'standard' employment in normal part-time situations, and are less likely to receive health insurance or a pension through their employer (indeed it has been argued that the growing cost of health insurance and pension contributions may encourage employers to recruit part-time and temporary workers). It has been suggested that atypical (and especially temporary) work is perceived as more normal in the USA highly mobile and flexible economy and labour market than in Japan or many EU countries.

No data is available on the content and extent of US collective bargaining on issues related to new forms of work, though there have been cases (according to NAFFE - see below) of trade unions using the collective bargaining process to fight 'two-tier' agreements, ensure equal pay and benefits for part-timers, limit the number of contingent jobs, and demand regular jobs for those contingent workers who want them. There is evidence of increasing interest in the issue in trade union circles. For example, AFL-CIO and a number of affiliated unions are members of the North American Alliance For Fair Employment (NAFFE), a 'network of organisations concerned about the growth of contingent work - including part-time jobs, temping, subcontracting - and its impact on the well-being of all workers', which calls for equal treatment (pay, benefits and protection under the law) regardless of employment status. NAFFE has campaigned for the rights of groups such as temporary agency workers and day labourers. Trade unions have been making efforts to organise contingent workers, but despite a few successes this has proved difficult. Other relevant trade union activities include opposition by building trade unions to Labor Ready, a temporary work agency which employs some 700,000 non-union skilled construction workers and manual labourers. The unions point to the fact that the company pays the minimum wage to most of its workers and provides no healthcare, sick pay, annual leave, life insurance, disability coverage or retirement plans.

There has also been some signs of interest in relevant issues in official circles. In 2000, the General Accounting Office (GAO) published a report that proposed the consideration of more portable and affordable benefits for some contingent workers. It also floated the ideas that temporary agencies provide some benefits themselves, or that contingent workers be offered tax incentives to purchase their own health insurance.

Conclusions

The information in this report has highlighted a number of major differences and a smaller number of similarities between the industrial relations systems in the EU, Japan and the USA, summarised below. However, the difficulties of comparing a relatively diverse grouping of 15 or 25 countries with two individual countries, mentioned earlier, should be borne in mind. While it is possible in many areas to speak of 'typical' or 'average' EU situations and use these for comparison with Japan and the USA, it should be stressed that a number of Member States fall outside these situations. For example, the UK and some new Member States arguably have as much or more in common with the USA in some industrial relations areas than they do with the EU 'norm'.

  • There is no typical 'EU model' of trade union organisation and structure that can be compared with Japan and the USA. In the current and new Member States, there may be a single significant national union centre/confederation - as in the USA - but more commonly multiple ones - as in Japan - either competing for the same potential members or organising different categories of workers. However, trade unionism at European level is unified and coherent, because of widespread membership of ETUC and its sectoral European industry federations. There is a degree of convergence between the EU, Japan and the USA in the form of a common trend towards consolidation and merger of trade union organisations.
  • A common feature is that union membership has generally been falling across the current EU, Japan and USA in recent decades. Although there are exceptions among the 'old' Member States, the overall trend is one of gradual steady decline, except in some of the new CEE Member States, where membership has tumbled rapidly. Union density has thus fallen quite steadily in Japan, the USA and almost all European countries. However, despite this shared trend, a major point of difference is that in the current EU, the average density (unweighted for the different sizes of the 15 countries), at 43%, is still more than double that in Japan (20%) and more than three times that in the USA (13%), and density remains at 50% or more in five Member States (though it is below 20% in France and Spain). Weighting the density figures gives an average for the current EU of 30%, which is still half as high again as density in Japan and more than double that in the USA. Union density is generally lower in the new CEE EU Member States than in the 'old' 15 (though two-thirds or more in Cyprus and Malta), with an unweighted average for all acceding countries of 34% and a weighted average of 22%, which is near Japanese levels. Unweighted average density in the expanded EU of 25 will fall to 39% and the unweighted average to 29% - still significantly higher that Japanese and US levels.
  • There are some shared features between the current EU and Japan in terms of the national intersectoral organisation of employers, with Japan and most (though not all) present EU Member States essentially having a single umbrella organisation representing private sector companies' employer and business/trade interests (though this is a recent development in Japan). This is also essentially the case at overall EU level, through UNICE (though with separate SME and public sector organisations, as in some Member States). The picture is more varied in the new Member States, though some follow the 'typical' current EU model. Some form of intersectoral bargaining with trade unions (though of a varying nature) is part of the remit of central employer bodies in 10 of the current EU countries and a few of the acceding countries (as well as at EU level), and they have close cooperative relations with trade unions, which may lead to joint texts or approaches, in three more present Member States. Japan has similarities with this latter group of three countries, with Nippon Keidanren not participating in bargaining, but involved in formal tripartite dialogue with government and unions - also common in the current and new EU Member States - and less formal bipartite dialogue with unions (while exerting some influence on company-level bargaining). The USA is very much at odds with Japan and most present and new EU Member States in having no identifiable national intersectoral employer body with an industrial relations role.
  • A major difference between the majority (12) of current EU Member States on one hand, and Japan, the USA and most new Member States on the other, is the strong presence of sectoral employer organisations with a collective bargaining role. Such bodies are rare in the new Member States (with three exceptions) and the USA, and non-existent in Japan (though employer associations may play a 'behind the scenes' role in coordinating member companies' bargaining).
  • The level at which collective bargaining is conducted is a fundamental difference in industrial relations between most current EU Member States on one side, and Japan and the USA on the other. In general, bargaining in the present EU is considerably more centralised than in its two competitor countries. Taking the example of pay bargaining, in only one of the current 15 Member States is the company very clearly the dominant bargaining level, and the sector is a dominant or important level of pay bargaining in 11 cases, while the intersectoral level is dominant or important in four Member States. By contrast, in both Japan and the USA, the predominant bargaining level for pay and all other issues is the individual company. Intersectoral or sectoral bargaining is virtually unknown. However, collective bargaining is generally much more decentralised in the new Member States than in the current EU. The company is the dominant pay bargaining level in seven out of 10 acceding countries. Sectoral wage bargaining has a dominant or important role in only four new Member States and this applies to the intersectoral level in only one case. The impact of EU enlargement will arguably be to shift the 'average' level of bargaining towards the company level and thus nearer the situation in Japan and the USA. Nevertheless, the intersectoral or sectoral level will remain the most important level of wage bargaining in at least 60% of the countries of the extended EU, including four of the five largest economies. Bargaining will thus remain much more centralised in the EU than in its two major competitors.
  • Collective bargaining coverage in the current EU averages 77% (unweighted for the relative size of countries) and 72% (weighted). This is nearly four times the rate in Japan (20%) and five times the rate in the USA (15%) and bargaining coverage thus arguably constitutes the single most marked difference between the EU and its two competitors. While Luxembourg and especially the UK have relatively low rates, the rate is 90% or more in France, Belgium, Sweden, Finland and Italy. In general, bargaining coverage is lower in the new CEE Member States, though relatively high in Cyprus and Malta, with an average rate for the acceding countries of 42% (unweighted) and 37% (weighted) - both figures are only a little over half of the rates in the current EU. The enlargement of the EU will reduce its overall unweighted coverage rate substantially, from 77% to 63%, but will cut the weighted average only from 72% to 67% - still over three times that found in Japan and over four times that in the USA. A further difference is that bargaining coverage in the current EU seems relatively stable - though the picture is more mixed in the new CEE Member States - while in Japan and the USA, it has been in more or less constant decline over the past 20 years or more.
  • Employee involvement is a further important area of difference between the EU and its two competitors. Of the current EU Member States, 13 have a widespread system of indirect or representational employee involvement at company/workplace level, usually through works councils or similar bodies, based on law or collective agreement. These systems give employee representatives defined rights to information, consultation and sometimes other forms of involvement. Ireland and the UK are exceptions, with statutory information and consultation limited to a number of specific issues, but this situation is set to change through implementation of the 2002 EU Directive on informing and consulting employees. At least half of the new Member States already have statutory works council-type bodies and all will implement the information and consultation Directive. A majority of current EU countries, and some acceding countries, have a statutory system of employee representation of company boards. At EU level, legislation has introduced EWCs for pan-European information and consultation in multinational companies, with around 700 having been set up so far. By contrast, neither Japan nor the USA have legislation providing for works council-type structures or board-level employee representation. In Japan, however, many companies have representational employee involvement through 'labour-management consultation organisations' dealing with issues quite similar to those handled by works councils. Many US companies have some form of employee involvement arrangements in place, but these tend to be limited in scope and appear in many cases to involve direct rather than representational participation.
  • Another key point of industrial relations difference between the EU and many of its Member States on one hand, and Japan and the USA on the other, appears to be the nature and extent of employment and labour law. At both EU level and in many Member States, there is detailed legislation regulating many areas of industrial relations, employment conditions and workers' rights. This is added to frequently as situations and priorities change. In Japan and particularly the USA, only the basic 'rules of the game' and/or a number of minimum rules on employment conditions are laid down in law. However, Japan seems to have been introducing rather more legislative regulation in recent years.

Turning from structures and processes to outcomes, a comparison of recent bargaining and developments relating to pay and working time highlights the following points.

  • Pay is, unsurprisingly, a central issue in collective bargaining in the EU, Japan and the USA. However, the much higher bargaining coverage in the current EU generally, and the key role of sectoral and intersectoral bargaining in many countries, means that collectively-agreed pay increases have a far wider significance then they do in the low-coverage, company-based systems of the USA and Japan. Nominal collectively agreed pay increases were highest in the USA in 2001 (at 3.9%), followed closely by the current EU (3.7%) and at a greater distance by Japan (2.0%). The same ranking was maintained in 2002 though the USA (3.9%) increased its lead over the current EU (3.4%) and Japan (1.7%). Figures are not available for all new Member States, but pay increases were generally above the EU 15 average in both years. Adjusting the figures for inflation, the real agreed pay increase was highest in Japan in 2001 and 2002, at 2.6% in both years. In the current EU, the average real agreed increase fell to 1.3% in 2002 from 1.4% in 2001, while in the USA it rose to 2.3% from 1.1%. The differences in agreed pay increase obviously reflect varying economic circumstances, but also in some cases explicit social partner policies. In the EU 15, there was a relatively widespread tendency towards increased pay moderation in 2002 (though with exceptions such as Germany), and in some countries this was based on tripartite or bipartite agreements. In Japan, the poor economic conditions meant that safeguarding jobs largely took precedence over pay increases in 2002, with unions moderating wage demands and in some cases even accepting cuts. In the USA, the issue of health insurance arguably replaced pay at the top of the bargaining agenda. Finally, one feature of pay which is common to the EU, Japan and USA is the persistent (if slightly narrowing) gender wage differential, with women earning less than men in all countries.
  • Working time reduction has been a major issue in collective bargaining in the current EU Member States over recent decades, though major agreed working time reductions have been largely absent over the past few years, with the notable exception of France. However, cuts are still being agreed in some countries in particular sectors and companies, often linked to the introduction of greater flexibility, while reductions seem high on the agenda in some new Member States. Working time also appears to be an important bargaining issue in Japan, though it seems to have resulted in a gradual reduction in hours over time, rather than any dramatic changes. Working time does not currently appear to be a high-profile bargaining theme in the USA. The average normal weekly working hours set by collective bargaining across the current EU stood at 38.2 hours in 2002, unchanged from the 2001 figure. Information is not available on average agreed weekly hours in the new Member States, but it is generally appears above the average for the EU 15. Working time reductions seem high on the industrial relations agenda in some acceding countries, such as Hungary and Slovakia. The average agreed week of 38.2 hours in the EU 15 in 2002 can be compared with the Japanese 'average scheduled weekly working hours' of 39.4 hours for full-time workers in 2002 ( a 0.2-hour increase on 2001). No relevant data are available on collectively agreed working time in the USA. All EU Member States and Japan have a statutory maximum or normal working week, though USA does not.
  • Average normal agreed or scheduled annual working hours (though definitions vary) appear to be longest in Japan, followed by the USA and the current EU (on average). These annual figures are influenced by the fact that paid annual leave is considerably higher on average in the current EU than in Japan or the USA. In 2002, the average number of days of collective agreed annual leave in the EU stood at 25.9, while the average paid holiday entitlement stood at 18.1 days in Japan (though less than half of this leave is taken on average). Even after 20 years' service, the average number of annual vacation days in the USA stands at only 20.3 days (1997 figure). The current EU Member States all have a statutory minimum period of paid annual leave (averaging 22 days), as does Japan (10 days after six months' service), but not the USA.

With regard to how the various industrial relations systems handle the current wave of company restructuring affecting the global economy, the following observations can be made (focusing in the case of the EU largely on the 15 current Member States).

  • Company restructuring has clearly been one of the most high-profile topics in industrial relations in the current EU in recent years. It has affected all Member States (though to differing extents and in varying ways, arguably partially reflecting specific systems of corporate governance) and there have been numerous high-profile and controversial cases. This has led to a number of EU and national initiatives on ways of managing such change. Similarly, in Japan, restructuring and the 'employment adjustments' which often accompany it have become an increasingly important industrial relations issue in recent years, following a long period of economic growth during which employment stability and 'lifetime employment' were the norm. Companies' traditional reluctance to make workforce reductions has weakened and various legislative measures have been taken in response (see next point), while maintaining employment has come to top the bargaining agenda. The USA differs from the EU and Japan in that its continuing major corporate restructuring does not appear to have achieved the same prominence in industrial relations. Constant restructuring appears to be widely perceived as normal - and indeed in some circles as positively beneficial - though recent trends towards relocation outside the USA and 'global outsourcing' have proved more controversial.
  • The EU and its present Member States have for many years had a legislative framework in place which regulates - mainly procedurally - a number of employment aspects of restructuring, applying in circumstances such as collective redundancies, business transfers and insolvencies. These laws lay down certain rules on employers' obligations and employees' rights (including information and consultation). There are also regulations providing for employee information and consultation rights on company restructuring-related matters at both European and national level. Japan formerly had little in the way of such regulation, but this has changed against the backdrop of difficult economic conditions and increasing job losses. New legislation has been introduced, notably regulating dismissals for the first time. Now, the law states that a dismissal that has no objective rationale or cannot be judged socially acceptable is an abuse of the employer's right to dismiss employees and thus invalid. Another example is a new legal requirement for employers making a large number of redundancies to adopt a plan to support re-employment systematically ahead of termination. In this area, the EU, and now Japan to some extent, differ considerably from the USA, where companies are by comparison almost free to restructure and dismiss workers at will (though larger companies must give 60 days' notice of plant closures or mass redundancies).
  • In all current EU Member States, certain aspects of company restructuring are subject to the statutory involvement of employee representatives (including trade unions in some cases). These aspects include collective redundancies and business transfers in all cases, plus in most countries a variety of workplace-related restructuring issues and wider company restructuring matters. Employees can influence restructuring through rights based systems of co-determination or 'social concertation' and/or through trade unions and collective bargaining. Furthermore, at European level company restructuring is often a major issue dealt with by EWCs, though with varying degrees of effectiveness. Overall, the importance of some form of employee involvement in restructuring seems to be widely accepted by all main actors in the EU. Japan is relatively close to the EU on this point, as restructuring is a key issue in company-level collective bargaining and consultation, though without a statutory basis. 'Corporate restructuring and/or reductions in business divisions' is an increasingly important theme in discussions between trade unions and management, and negotiations tend to be initiated at a comparatively early stage in the restructuring process. In terms of other forms of employee involvement, workforce reduction issues are dealt with by a majority of labour-management consultative bodies. Again it is the USA which stands alone, with no legislation requiring employers to inform (though see previous point), consult, or negotiate with employees over restructuring, while only a few collective agreements contain agreed procedures related to company restructuring (though many contain provisions designed to enhance workers' employment security).

Finally the treatment of 'new' or 'atypical' forms of work also clearly illustrates some differences and convergences between industrial relations systems.

  • A now familiar pattern applies to the legislative regulation of new forms of work. There is a high level of specific regulation in the EU, a lower but increasing level in Japan, and virtually none in the USA. At EU level, recent Directives have laid down rules - essentially aimed at ensuring equal treatment for atypical workers - on part-time work and fixed-term contracts (while a draft on temporary agency work is under discussion), and the EU-level social partners have concluded an agreement on telework. All current Member States (and increasingly the new ones) regulate various aspects of new forms of work. While there has been a widespread tendency to promote and permit various new forms of work, this has generally been accompanied by the establishment of a number of rules for their operation. Rules have been relaxed in some Member States, but there has rarely been total deregulation. In Japan, some aspects of atypical work have been subject to legislative regulation despite the traditional model of regular or lifetime employment. As employment patterns have diversified, this regulation has increased, with laws on certain aspects of part-time work and the conditions of employment for temporary agency workers (accompanying a deregulation of temporary work agencies). However, legislation does not provide for equal treatment of atypical workers, though the issue of some specific form of balanced treatment is currently under discussion with regard to part-timers. The USA appears to have no specific legislation to regulate any new forms of employment, with employment law applying to all workers within a context of a generally low level of regulation of any employment matters.
  • Collective bargaining plays a role in regulating new forms of work in most current EU Member States. Aspects of part-time and fixed-term work are covered by agreements at various levels in many countries, and there is an increasing trend towards collective bargaining relating to temporary agency work and teleworking. Unfortunately, there are no data on the extent of bargaining on these matters in Japan and the USA, but in neither case does it appear to be extensive.

Mark Carley, SPIRE Associates

Page last updated: 24 February, 2004
About this document
  • ID: TN0401101F
  • Author: Mark Carley, SPIRE Associates
  • Language: EN
  • Publication date: 24-02-2004