Industrial relations in the EU, Japan and USA, 2003-4

This overview examines industrial relations developments in the EU, Japan and USA over 2003-4, and focuses on the topical issues of migration and pensions.

Since 2000, the European Foundation for the Improvement of Living and Working Conditions has each year conducted a project comparing aspects of industrial relations in the European Union and the world's two other largest economies - Japan and the USA. The essential aim is to provide a picture of similarities and differences in both basic structures and current developments, not least to help illuminate what does and does not make the 'European social model' distinctive, and to assist with 'benchmarking' the EU against its major competitors in an era of deepening economic globalisation.

Earlier comparative reports from the project (TN0111148F, TN0212101F and TN0401101F) have examined in some detail basic industrial relations actors and processes such as collective bargaining, trade unions, employer organisations, labour legislation and employee involvement. Readers are referred to these reports for background information. The current report looks at the main developments over 2003 and much of 2004 in these areas before going on to examine how the various industrial relations systems deal with two very topical issues - the increasing level of international migration (especially labour migration) and pensions provision (particularly occupational pensions) at a time of demographic change.

The information in this report on the EU is taken mainly from the European Industrial Relations Observatory (EIRO). In the areas of migration and pensions especially, there are some gaps in the information available for the 10 new Member States that joined the Union in May 2004. The data for Japan and the USA are taken largely from special reports on industrial relations developments in 2003-4 prepared for the European Foundation for the Improvement of Living and Working Conditions by experts in these countries - see JP0411101F and US0411101F- 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 Foundation conference on 'Industrial relations in the EU, Japan and USA' held in Dublin on 18-19 October 2004.

This conference was attended by experts from the countries directly concerned (the USA, Japan, and various EU Member States) and international organisations. Furthermore in 2004, the Foundation aimed to bring in a wider perspective, and also invited experts to present information on developments in Asia and Latin America. According to the Foundation, this approach nevertheless raised methodological questions, as comparisons are difficult between individual countries - USA and Japan - and a structure such as the EU, or a subcontinent such as Latin America. Diversity is also huge among Asian countries, as they are significant in number, with very varied cultures and religions. Furthermore the importance of the informal sector and the different role of social actors in Latin America and Asia raises questions at a conceptual level. The Foundation believes that if it wants to develop meaningful monitoring of industrial relations and labour market evolutions beyond the OECD, these challenges will certainly have to be addressed in the future.

Economic and employment context

The economic and labour market contexts for industrial relations in EU, Japan and the USA are very different - see table 1 below. The EU economy slowed over the 2001-3 period, though with some signs of renewed growth in 2004. The 'old' 2004 EU15 experienced lower growth over 2001-4 (0.7% in 2003 and 1.9% predicted for 2004) than the 10 new Member States that joined in May 2004. After being largely stagnant over 2001-2, the Japanese economy started to recover in 2003-4. The US economy, however, experienced a rising trend over the whole 2002-4 period.

EU inflation has been falling slowly since 2001 (when it stood at 2.5%), and was projected to reach 1.9% in 2004. The EU15 experienced the same trend, falling from 2.2% in 2001 to a predicted 1.8% in 2004. In the USA, inflation has been fluctuating on a yearly basis, falling in 2002, rising in 2003 and falling again in 2004 (to a projected 1.4%, somewhat below the EU level). In Japan, prices have actually been falling since 1999. However, in 2003 the fall slowed and in 2004 prices are predicted to have been stable.

In the EU, employment levels rose very slightly in 2003 (by 0.2%, the same level as in the EU15), as they had the previous year. After slight falls in 2001 and 2002, employment again grew in the USA in 2003 - and more strongly (at 0.9%) than in the EU. The picture remains very different in Japan, where 2003 saw a continuation of the decline in employment witnessed since the late 1990s. The Japanese employment rate, however, was 5.5 percentage points above that of the EU, at 68.4% in 2003 - the EU15 was slightly closer, at 64.3%. The USA still outstrips its rivals in this area, at 71.2%. In all three cases, male employment rates exceed female rates.

Unemployment levels in the EU remain much higher than in Japan and the USA. The EU's projected rate for 2004 is 9.1%, compared with 5.5% in the USA and 4.8% in Japan. The accession of the new Member States has pushed up overall EU rate - the 2004 figure for the EU15 alone is 8.1%, though this is still substantially higher than in Japan and the USA. The EU rate rose in 2002 (to 8.9%) and 2003 (to 9.1%), before flattening out in 2004 (the same trend was observed in the EU15). The US rate also rose in 2002 (to 5.8%) and 2003 (to 6.0%), but is projected to fall quite sharply in 2004. Japan's long-term upward trend peaked in 2002 (at 5.4%), before falling slightly in 2003 and more substantially in 2004.

Nominal pay increases per employee in the EU exceeded those in Japan and the USA in 2003, as they had in 2002 and 2001. The same was true of the EU15, though here increases (3.0% in 2003) were somewhat lower than in the extended EU. The rate of increase fell over 2001-3 in the EU and USA, while in Japan, where nominal pay has been falling for several years, the rate of decline sharpened in 2002 before lessening in 2003. The projections for 2004 indicate a slight fall in the rate of increase in the EU (and EU15). However, a sharp rise in the US figure brings its nominal pay increase above the EU figure for the first time since 2000, while 2004 should also see Japan's first nominal increase since 2000. Adjusting nominal compensation increases for inflation, in 2003 real increases were greatest in Japan, followed by the EU (0.6% in the EU15) and the USA, while in 2004, the USA is projected to take over the leading position, followed closely by Japan and at some distance by the EU (the EU15 figure is the same as for the extended EU). The rate of increase in labour productivity rose in all cases from 2003 to 2004, with the USA achieving the greatest rise and Japan the least - in 2004, US productivity is projected to increase by around twice the EU rate (with the EU15 increase slightly lower, at 1.6%). In 2003, nominal unit labour costs grew in the EU (with the EU15 showing the same 2.2% growth) and remained almost static in USA, while they fell in Japan. In 2004, the rate of increase is projected to slow in the EU (and EU15) and remain virtually unchanged in the USA, with Japan's downward trend slowing. Finally, adjusting the nominal unit labour cost increases for inflation, they were static in the EU in 2003, but fell in 2004. Real unit labour costs fell in both 2003 and 2004 in Japan, as they did - more sharply - in the USA.

Table 1. Economic and employment indicators, EU, Japan and the USA, 2003-4
. EU25 Japan USA
2003 2004 2003 2004 2003 2004
GDP growth 0.8 2.0 2.5 3.4 3.1 4.2
Inflation (CPI) 2.0 1.9 -0.3 0.0 2.3 1.4
Employment rate 62.9 - 68.4 - 71.2 -
Employment growth 0.2 - -0.2 - 0.9 -
Unemployment rate (Eurostat definition) 9.1 9.1 5.3 4.8 6.0 5.5
Growth in nominal compensation per employee 3.1 3.0 -0.7 1.1 2.3 3.3
Growth in real compensation per employee (private consumption deflator) 0.9 0.9 1.9 2.1 0.6 2.2
Labour productivity growth 0.9 1.7 2.7 3.0 2.2 3.3
Nominal unit labour cost growth 2.2 1.3 -3.3 -1.8 0.1 0.0
Real unit labour cost growth 0.0 -0.8 -0.8 -0.8 -1.6 -1.0

Sources: European Commission - Employment in Europe 2004 and Statistical Annex to European Economy, autumn 2004. Note: 2004 figures are projections.

Collective bargaining

Bargaining levels and structures

In 2003-4, collective bargaining continued its long-term decline as a means of setting the pay and employment conditions of workers in Japan and the USA, mirroring falling trade union membership. In Japan, with bargaining conducted exclusively at company level and no mechanisms for the extension of agreements beyond the signatories, bargaining coverage exactly matches union density. The latter fell from 20.2% in 2002 to 19.6% in 2003, continuing a downward trend that has been in progress for nearly 30 years. However, it should be noted 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 - though not always being directly reflected in average wages. In the USA, the coverage of bargaining (basically conducted at company level) is similarly restricted to trade union members - who fell from 13.2% of the workforce in 2002 to 12.9% in 2003 and 12.5% in 2004. However, 13.8% of the workforce were represented at the bargaining table in 2004, with the addition of workers were organised in a 'union-typical' manner, such as members of certain employee associations, and those who report no union affiliation but whose jobs are covered by a union or an employee association contract.

The bargaining coverage picture in the EU is mixed and less easy to measure on a year-by-year basis - not least because there is not such a close correlation with union membership as there is in Japan and the USA. On an individual country basis, there was little change reported in 2003-4, but at overall EU level the accession of the 10 new Member States in May 2004 will have reduced somewhat the coverage of the EU workforce by bargaining. Though with major exceptions in both groups (eg relatively low coverage levels in the UK, and relatively high coverage in Cyprus), collective bargaining coverage is much lower in the new Member States, covering somewhere between a third and four out of 10 of the workforce on average, compared with around three-quarters in the old EU. The expanded EU's overall coverage rate has, as a result, been cut - by around five percentage points on a weighted basis, according to EIRO calculations in 2003 (TN0401101F). However, it should be noted that somewhere around two-thirds of the workforce of the expanded EU are covered by collective bargaining - a rate over three times that in Japan and over four times that in the USA.

As usual, collective bargaining in 2003-4 occurred almost exclusively at individual company or workplace in Japan and the USA (though with a few exceptions in the latter case, such as a multi-employer agreement for Pacific coast dockworkers). 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 small and medium-sized enterprises (SMEs). Industry-level union federations establish industry-wide targets and work out specific strategies in order to improve the negotiating capabilities of their affiliated enterprise unions. However, since the late 1990s lower economic growth and more intensive international competition have increased the degree of disparity in performance between companies, thus making it more difficult for unions to set industry-wide wage increase targets and to achieve 'horizontally egalitarian' pay rises. Bargaining in the USA has become increasingly decentralised in recent decades, with the level shifting, first from multi-employer pattern bargaining (whereby an agreement is achieved in a particular company or companies and then extended to other firms in the sector) and 'master' agreements to the company level, and then from the company level to the individual worksite.

In the EU countries, bargaining in 2003-4 was conducted at all levels from the national intersectoral to the individual workplace, depending on national systems. That the central intersectoral level plays a key role in many countries was underlined by the conclusion, renewal or negotiation of major comprehensive national framework agreements on pay and other conditions in countries such as Belgium, Finland, Greece, Ireland, the Netherlands, Slovenia and Spain, and intersectoral agreements on specific themes in countries such as Estonia, France, Hungary, Italy and Luxembourg. The sector is the dominant level of bargaining on pay and conditions, or an important level, in a majority of the old 15 EU Member States (the main exceptions being Ireland, Luxembourg and the UK) and 2003-4 saw the bargaining process (whether annual or multiannual) in these countries proceeding more or less as usual. However, there are signs of change at various levels - see box 1 below.

Box 1. Decentralisation of bargaining in the EU The new EU Member States generally have much more decentralised collective bargaining systems than the 'old' EU. The company is the key bargaining level (though, because of low bargaining coverage rates in many countries, this does not mean that it covers a high proportion of all companies and employees) in all 10 countries apart from Slovakia, Slovenia and Cyprus. Sectoral bargaining (in a broad sense) plays the dominant role only in Slovakia and Cyprus, is a very important bargaining level in Slovenia, and relatively significant bargaining in Hungary and to a lesser extent the Czech Republic. EU enlargement has thus arguably shifted the 'average' level of bargaining towards the company level. Change is also stirring in a number of individual countries. Most notably, there is pressure towards decentralisation in a number of old EU Member States that have long-standing systems of sectoral collective bargaining (though generally the move is towards 'organised' decentralisation while maintaining a sectoral framework). In Germany, to take a leading example, there have been numerous calls from employers and opposition politicians to decentralise the bargaining system and allow for more firm-level deviations from sectoral agreements (DE0312202F). Thus in the 2004 bargaining round trade unions were confronted with demands from employers to extend flexibility (notably on working time) and to allow the parties at company level to deviate from sectoral agreements. Unions managed only partly to resist these demands, and in a number of cases saw themselves forced to concede 'opening clauses' (DE0408204F). Another example is France, where new legislation passed in 2004 overturned the previous hierarchy of collectively agreed norms, introducing the possibility of company-level agreements departing from sector-level agreements in a way that is unfavourable to employees, in some circumstances (FR0404105F). A final example is Italy, where debate in 2004 over reform of the current two-tier bargaining system - whereby company/local agreements are concluded within the framework of sectoral agreements, with each level dealing with specific issues - has raised the possibility of giving greater weight to the decentralised level (IT0412306F).


As ever, pay was at the heart of collective bargaining in the EU, Japan and USA in 2003-4. In Japan, given difficult economic conditions, the spring wage negotiations have been difficult for trade unions over the past few years as they have been forced to choose between maintenance of employment levels and pay rises. Furthermore, since 1999 in particular, employers have been seeking to prioritise lump-sum bonuses and incentives over 'base-ups' (pay revisions that reflect an upward trend in the prices index or the company's business performance) and 'regular' pay increases (wage rises based on the individual employee’s ability, age or length of service). In 2002 (JP0311101F), the Japanese Trade Union Confederation (Rengo) discontinued the submission of a unified demand for a basic wage rise. With the majority of enterprise unions obliged to defer their demands for a 'base-up' increase, the primary focus of wage talks has been on continuation of regular pay increases, and the amounts of lump-sum bonus payments. The spring wage negotiations pushed up the average wage level by 1.63% in 2003 (the 2002 level was 1.66%) and 1.67% in 2004 (as noted above, the wage increases that result from the shunto, while having a major influence on general wage levels, are not always directly reflected in average wages). As for bonuses, which account for a large proportion (probably a quarter, on average) of the annual income for Japanese workers, workers at major companies achieved a 3% rise for the 2003 summer bonus, after a fall of 4.3% in 2002.

Rengo unions have recently been placing special emphasis on achieving wage increases for workers in SMEs, and Rengo established its first-ever unified wage rise demand for member unions in such companies in the light of a growing gap with larger corporations. According to a Rengo analysis, the 2004 spring wage negotiations led to larger wage hike in SMEs than in 2003 (when the increase stood at 1.17%). Furthermore, since 2002, Rengo has been began demanding an additional hourly wage rise for part-time workers in the spring wage negotiations, with the aim of closing the wage gap with full-timers. In the 2004 spring wage talks, many trade unions achieved JPY 10 - JPY 15 hourly wage increases for part-timers, which is considered a modest achievement.

Bargaining plays a far less important role in setting overall national pay increases in the USA 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 27% higher than the latter's in 2003. Negotiated wage increases averaged 3.4% in 2003 and 3.5% in the first half of 2004. This indicates that economic growth and major productivity gains have not yet translated into improved wage settlements. Negotiated wage increases had been higher during the recession period - 3.9% in both 2001 and 2002, at a time of major job losses in manufacturing and slower economic growth. Pay settlements in manufacturing lagged behind construction and the state and local government sectors by about 0.5 percentage points in both 2003 and 2004.

Collective bargaining plays a key part in overall pay determination in most old EU Member States, and a number of the new ones. The nature of bargaining's role in pay setting differs widely between the countries - the different bargaining levels (intersectoral, sectoral, company etc) play different parts, while the importance of bargaining in pay determination differs considerably between sectors of the economy and groups of workers. According to EIRO estimates (TN0303102U), average collectively agreed pay increases across the whole EU stood at 4.0% in 2003, down from 4.4% in 2002 - no data are yet available for 2004 - though with very considerable differences between countries. The new Member States saw average increases of 6.5% in 2002 and 5.8% in 2003, while the average EU15 rises were 3.3% in 2002 and 3.0% in 2004 (there were also major variations within the two groups of countries).

In 2003, there was a widespread downward trend in nominal wage increases across the EU15, with only Finland, Spain, Italy and the UK recording a higher increase than in 2002. In many cases, those involved in bargaining sought to concentrate on safeguarding and creating employment in a difficult economic climate, resulting in pay moderation. National agreements explicitly sought pay moderation in countries such as the Netherlands (NL0310103F) and Belgium (BE0302302F). In the countries then awaiting EU accession, the average increase in 2003 was still substantially higher than that in the current EU, but the differential fell between 2002 and 2003 (with moderation in countries such as Estonia, Hungary and Slovenia), suggesting that pay trends in the new Member States may be converging downwards towards those in the old EU.

Full data are not yet available on the outcomes of pay bargaining in the EU in 2004. However, the indications are that there was a widespread mood of restraint. In Germany, collective agreements concluded in the first half of 2004 (affecting about 27% of all employees covered by collective agreements) increased pay by around 2.0% in 2004, which is below the average pay increase of 2.5% in 2003 (DE0408204F). The major private sector settlement in Denmark focused mainly on improvements in areas such as occupational pensions, paid parental leave and sick pay, rather than on substantial pay increases (DK0405102F). In Ireland, the second half of the current three-year national agreement - negotiated in 2004 - provides for average pay increases of 5.5% over 18 months, compared with 7% over the first 18 months (IE0409203F). New three-year agreements signed in most sectors in Sweden provided for relatively modest pay rises of under 7% over three years (SE0403103F). Strong calls for moderation also marked the negotiations over new national agreements for 2005 and beyond in Belgium and Finland. However, although wage increases fell away sharply in 2004, the earlier national pay restraint agreement in the Netherlands was abandoned by trade unions in 2004 (NL0407101N) in opposition to government reform plans.

In some new Member States, it also appears that pay moderation was the rule in 2004. For example, in Slovenia, a new private sector pay policy agreement covering 2004-5 provides for increases such that the real growth in average gross pay per employee will lag behind productivity growth by at least one percentage point (SI0405103F). However, there were stirrings of wage militancy in some countries as economic conditions improved. Poland saw a wave of pay demands in late 2004 in sectors such as the metal-processing, coal-mining, petroleum, automotive, and food industries, with unions calling for workers to benefit from a recent improvement in the economic climate but employers warning that over-hasty spending of the fruits of economic growth may lead to a reduction of investments and thus endanger this growth (PL0411104F). 2004 also saw a number of strike and protests in private sector enterprises in Slovakia, focusing on unpaid wages and dissatisfaction with the pay outcomes of collective bargaining (SK0410101N). This is unusual in a country where strikes have been rare and trade union protests largely confined to the public sector and responses to government policies.

Nominal collectively agreed pay increases were highest in the expanded EU in both 2002 and 2003, followed by the USA, the EU15 and Japan - see table 2 below. In 2004, the increase in the USA was more than double that for Japan (EU figures not yet available). Adjusting for inflation, real agreed pay increases were highest in Japan in both 2002 and 2003. In 2002, the USA was in second place, followed by the expanded EU and then the EU15. In 2003, the expanded EU followed Japan, above the EU15 and USA (which had the same increase). Finally, in 2004 real agreed increases in the USA outstripped those in Japan.

Gender wage differentials persist in the EU, Japan and the USA. In the expanded EU (including two candidate countries), women's earnings (based on hourly figures) were on average 81.7% of men's in 2003 (according to EIRO estimates TN0403103U) - 81.4% in the EU15 (plus Norway) and 82.3% in the new Member States/candidate countries. Overall, the gap seems to be narrowing slightly. In the USA, the equivalent figure for 2003 was 76%, down from 77% in 2002, the first increase in the gap since 1995. In Japan, the average scheduled cash earnings for female full-time workers stood at 66.8% of male full-time workers’ earnings in 2003 (66.5% in 2002), showing a gradual upward trend.

Table 2. Collectively agreed pay increases in the EU, Japan and the USA, 2002-4 (%)
. Expanded EU* EU15 Japan USA
2002 2003 2004 2002 2003 2004 2002 2003 2004 2002 2003 2004
Average nominal collectively agreed pay increase 4.4 4.0 - 3.3 3.0 - 1.7 1.6 1.7 3.9 3.4 3.5
Average real collectively agreed pay increase 1.6 1.4 - 0.7 1.1 - 2.6 1.9 1.7 2.3 1.1 2.1

Sources: EIRO, Eurostat, JILPT, Bureau of National Affairs. See main text for definitions. * Minus Czech Republic and Lithuania.

Working time

In 2003-4, working time did not appear to be a significant issue on the US collective bargaining agenda (though it was an important topic in legislative terms - see below under 'Legislative developments'), but appeared to gain importance to some extent in Japan. The matter is perennially important in bargaining in many EU countries, and in some of these there have been signs over the past two years of a change of tack.

Working hours are a major theme for company-level bargaining in Japan. Since discontinuing the submission of a unified demand for a basic wage rise in 2002 (see above), the Rengo trade union confederation has placed more emphasis on improving common 'institutional' frameworks for all workers, including reduced working hours and work-sharing schemes. In the 2004 spring bargaining round, one of Rengo's priority issues (though without any notable success) was eradicating unpaid overtime work, which is an increasing phenomenon. Furthermore, in 2001 Rengo and the then Nikkeiren employer organisation concluded a 'social agreement on employment', which affirmed that employers would make comprehensive efforts to maintain employment and that, in return, employees would accept shorter working hours and the resulting lower wages (JP0211101F), an approach known as work-sharing. The agreement would appear to have led to some diversification of working patterns. In 2003, 'average scheduled weekly working hours' in Japanese companies - 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, the same level as in 2002.

For several decades at least, the reduction of the duration of working time (or at least the normal hours set by collective agreements) has been a major feature of western European industrial relations - arguably, to a far greater extent than has been the case in Japan and the USA. During the 1980s and 1990s, many European countries saw major reductions in working time, though occurring at different paces and in different ways. In the past few years, dramatic reductions have been rare - the main exception being the introduction by legislation of a 35-hour normal working week in France over 2000-2 - but more gradual cuts have continued in many countries. EIRO started producing a rough average for normal collectively agreed working hours in western European countries in 1999, when the figure stood at 38.6 hours. In 2003, it had fallen to 38.0 hours - the average has thus continued to creep downwards, albeit slowly. Against this backdrop, in the EU15 flexibility, in various forms, has become the main theme in negotiations on working time (sometimes in exchange for small-scale reductions). For example, sectoral agreements introducing or expanding working time flexibility were signed over 2003-4 in countries such as Denmark, France, Germany, Italy and Spain.

In the new Member States, the active role of collective bargaining in setting normal weekly hours is relatively slight, especially in many central and eastern European countries, and collective agreements often do not tend to deviate from the statutory normal hours (usually 40 hours). The average agreed weekly hours across the 10 countries stood at 39.7 in 2003, down from 39.8 in 2002. In general, the new Member States still have a 40-hour normal week, with the exceptions of Slovakia and Cyprus, where agreed hours are around the EU15 average. The 40-hour week is relatively new in some countries, and overall there also appears to be a downward trend across the new Member States. However, in general, working time reductions do not appear to be high on the agenda at present, though with some exceptions, such as Hungary (HU0307101N).

Some events in 2003-4 suggest that Europe’s downward trend in working time may be turning - see box 2 below.

Box 2. Signs of an end to working time reductions in the EU? In Germany in 2003, the IG Metall trade union failed in an attempt to extend the 35-hour week to eastern German metalworking (DE0307204F). Still in German metalworking, a new industry-wide collective agreement signed in February 2004 extended existing provisions allowing company-level flexibility (DE0403203F). It allows deviations from sectorally agreed provisions, in areas including the extension and reduction of working time, with and without wage compensation, in certain cases, if this proves necessary to achieve a sustainable development of employment. Notably, the parties at company level can agree, in certain circumstances, that up to 50% of employees can work up to 40 hours a week. On the basis of these provisions in the sectoral agreement, in June 2004, IG Metall reached an agreement with the Siemens electronics group to increase weekly working hours from 35 to 40 at two mobile phone plants in Germany, with no extra pay for the workers involved (DE0407106F). The aim was to prevent jobs being 'exported' from Germany (to Hungary). In July 2004, DaimlerChrysler, the German-US motor manufacturer, announced an agreement that will safeguard 6,000 jobs in Germany until 2012 in exchange for concessions on working practices and planned pay increases, including a gradual increase in working time to 39 hours a week for services staff (DE0408102N). In addition, 20,000 employees in the development and planning departments may have their weekly working time increased from 35 to 40 hours by a supplementary agreement. It also emerged that agreements to lengthen weekly working time, in some cases without any additional pay and in exchange for no-redundancy commitments, were under discussion at a number of major companies across a range of industries in Germany. These cases produced enormous interest and debates about their meaning, both in Germany and more widely in Europe, and they were not completely isolated. For example, several similar agreements were reported in France. Notably, in July 2004, workers at the Bosch car components plant at Vénissieux ratified an agreement reached between management and two trade unions (FR0408101N). The deal aims to safeguard investment and jobs at the plant, in return for concessions on pay and working time. The current average weekly working time of 35 hours over the year will rise to 36 for manual workers, and a number of days off will be abolished. This deal proved controversial, especially in trade union circles, and highlighted debates about France’s 35-hour week legislation (FR0408108F). This law is coming under increasing pressure from the current conservative government and employers, and has already been watered down, in terms of relaxing its provisions on overtime work and suspending its application to small companies. These various events sparked a series of calls for an increase in agreed working time by firms and employer organisations in other countries - such as Austria (AT0407201N), Belgium (BE0408301N), the Netherlands (NL0408104F) and Slovenia (SI0411302F). Improving competitiveness was the main stated aim of such proposals. Trade unions reacted fiercely to the employers’ calls, which they denounced as opportunistic, and argued that a few specific company agreements, reflecting very particular conditions, cannot be generalised.

Other issues

Beyond pay, data on exactly what issues are covered in collective bargaining are often hard to come by, especially in countries where agreements are concluded primarily at company (or lower) level. Only in Japan and some EU countries, mainly those with sectorally-based bargaining systems, are relatively comprehensive data available on this issue

According to a large-scale official survey conducted in Japan in 2002, in the previous three years enterprise unions had conducted collective bargaining on pay in 58.1% of cases and working hours in 37.9% of cases. The most common other issues subject to bargaining were: employment and personnel affairs - 41.5% of cases; management policy - 22.0%; working environment - 20.9%; measures to secure employment - 20.3%; childcare and care leave - 19.3%; welfare - 16.6%; interpretation of collective agreements - 13.0%; health management - 11.1%; equal treatment of women and men - 10.8%; training - 9.8%; and working conditions of part-time and fixed-term workers - 7.7%. All of these issues, though, are dealt with more frequently by management and unions via consultation forums than collective bargaining proper. With regard to the regular annual spring bargaining round, there are recent signs of diversification. As noted above, since 2002 the Rengo union confederation has placed more emphasis on improving common 'institutional' frameworks for all workers, including reduced working hours, work-sharing schemes and extension of retirement age limits. In the 2004 spring wage negotiations, Rengo targeted three main issues: eradicating unpaid overtime work; improving working conditions for part-time workers; and providing better support to employees of SMEs. The Japan Business Federation (Nippon Keidanren) has been pressing for the spring wage negotiations to become a forum for a wider variety of discussions, including corporate personnel systems, rather than focusing exclusively on wage rise talks as in the past. It believes that the 2004 wage talks represented a positive move in this direction. An interesting development the 2004 spring wage negotiations was that some member unions of the Japanese Electrical Electronic and Information Union focused on 'proper work-family balance', and achieved better conditions on paternity leave in order to improve living standards for wage earners.

There is no official source of data on the contents of bargaining in the USA. However, in the past few years there seems little doubt that the single most important issue has been health insurance. Spiralling healthcare costs dictated the general tone of collective bargaining in 2003-4, thus continuing the pattern laid down in 2001-2 (US0311101F). In 2003, health costs rose by 12%, and were expected to go up another 13% in 2004. During 2003, in large collective agreements covering 1.5 million workers health costs were centre-stage and this usually involved 'cost-shifting' and some wage trade-offs (whereby unions agree to lower wage increases in order to preserve health and other benefits). Indeed, healthcare cost-shifting was at the centre of one of the USA's largest strikes in decades (see below under 'Industrial action'), and remains the most contentious bargaining issue. The number of US residents without health insurance coverage rose to 45 million by September 2004. Occupational pension provision was another major issue in bargaining (see below under 'Pensions and industrial relations').

In the EU25, the current bargaining agenda is far from heterogeneous. Notably, the agenda seems much wider in some of the old Member States but much more restricted - and often very much pay-centred - in many of the new ones. Issues that seem to have gained ground in many countries over recent years - though quantitative data are not available from most countries - include: occupational pensions (see below under 'Pensions and industrial relations'), equality and diversity issues, including reconciliation of work and family life; job security; the regulation of atypical employment; and training and skills development.

Legislative developments

A key aspect of industrial relations in the European Union is that, at overall EU level and in many individual Member States, there is legislation regulating in detailed way many areas of industrial relations, employment conditions and workers' rights, which is added to and amended 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. However, 2003-4 were relatively busy years in legislative terms in these two countries, and Japan especially seems to be increasing the volume and scope of its employment legislation to deal with new economic and social conditions.

In the USA, under the current Republican administration of President George Bush, one of the main items of basic employment legislation - the 1938 Fair Labor Standards Act (FLSA), which sets out rules on minimum wages and working hours - has been subject to amendment. The administration regards the FLSA as outdated and working against the interests of both employers seeking greater flexibility and workers seeking more control over work schedules. Under the FLSA, workers must be paid overtime pay (at a rate of time and one-half) for hours worked over 40 hours per week, except for overtime-exempt categories such as 'professional' or 'executive' workers. Despite opposition from trade unions and in Congress, the FLSA has been amended with effect from August 2004, reclassifying certain jobs as overtime-exempt and thus removing entitlement to overtime pay. Under the new rules, any workers who perform supervisory duties some of the time are no longer entitled to overtime pay rates, even if most of the time they perform non-supervisory work. The US Department of Labor initially claimed that roughly 750,000 workers would lose overtime protections under the new rules, but that others would be entitled to overtime pay because of adjustments in the minimum wage level that a worker needs to earn in order to be overtime-eligible. Critics claim that more than eight million workers will lose overtime entitlements. A further reform of the FLSA was announced in late 2004. The aim is to promote flexible working time arrangements, with compensatory time off for extra hours worked. The proposed legislation would allow an employee to work, for example, 50 hours per week without overtime pay, and to be compensated by working 30 hours the following week.

By its normal standards, Japan has seen hectic employment law activity in recent years - see box 3 below.

Box 3. New employment legislation in Japan in 2003-4 A Council for Regulatory Reform (CRR), established by Japan's current government, has called, in the employment/labour field, for legal reform to encourage labour mobility, industrial structure transformation and diversified working patterns. Following discussions in the Labour Policy Council (comprising representatives from the academic, labour and business communities) based on the CRR's philosophy, lawmakers in 2003 amended the Labour Standards Law, the Worker Dispatching Law and the Employment Security Law, all with effect from 2004. Due to fiscal difficulties, the Employment Insurance Law was also amended and came into effect in 2003. In 2004, parliament also amended the Law Concerning Stabilisation of Employment of Older Persons and enacted the Employment Tribunal Law.
  • The revised Labour Standards Law notably stipulates that in cases where a dismissal is not based on any objectively reasonable ground, and is not seen as a rightful decision in light of current 'social norms', the dismissal will be deemed invalid, constituting an abuse of the right to dismiss. It also increases the maximum duration of a fixed-term labour contract from one to three years (five years for workers engaged in highly specialised duties and those aged 60 and above) and seeks to enhance working time flexibility for workers engaged in planning and similar management-related tasks.
  • The revised Worker Dispatching Law relaxes the restrictions on the length of time a worker may be 'dispatched' and the types of jobs dispatched workers can engage in - 'worker dispatching business' is the term used in Japan to refer to temporary work agencies. Specifically, workers can now be dispatched to manufacturing sites, and the maximum period has been extended from one to three years. In addition, guidelines have been incorporated to encourage the transfer of dispatched workers to regular employee status.
  • The Employment Security Law grants local public bodies the right to provide free job placement services.
  • The revised Employment Insurance Law seeks to stabilise the unemployment insurance programme’s finances. The contribution rate will be raised to 1.6% from April 2005. Both the rate and maximum amount of unemployment benefits have been lowered, and a single system determining the period of payments to both former regular employees and part-time workers has been established to cope with diversification in employment patterns.
  • The new Employment Tribunal Law enables a panel in a district court (comprising a judge and two labour experts) to provide rapid, specific solutions for labour disputes, though the current labour mediation schemes, managed by public administrative organisations, will continue to play a central role in the settlement of labour disputes.
  • The revised Law Concerning Stabilisation of Employment of Older Persons requires employers to 'extend retirement age' or 're-employ' to secure the employment of their employees until the age of 65. The aim is to enable older people to work until their pensionable age.

At EU level, 2003-4 was a relatively quiet period in terms of the adoption of new employment and labour legislation. A Directive was adopted on the employee involvement arrangements to apply in the European Cooperative Society, a new (optional) form of European-scale cooperative business (EU0306201N). In the field of health and safety, Directives were adopted on the minimum requirements regarding the exposure of workers to the risks arising from physical agents (EU0212202N) and electromagnetic fields (EU0405202N). Discussions continued on a controversial draft Directive on the working conditions for temporary (agency) workers (EU0410204F). In October 2004, the EU-level central social partners signed a framework agreement on work-related stress, which aims to establish a framework within which employers and employee representatives can work together to prevent, identify and combat stress at work. The agreement will be implemented in accordance with the procedures and practices specific to individual countries rather than by an EU Directive (EU0410206F)

In terms of new EU legislative proposals, highlights in 2003-4 were draft Directives: amending the Directive concerning certain aspects of the organisation of working time (originally adopted in 1993), focusing on the 'opt-out' from the maximum average working week, the treatment of on-call working and reference periods for the calculation of maximum average working time (EU0410205F); on cross border mergers of companies, including provisions on what happens to existing systems of worker participation on the boards of companies involved in such mergers; and on minimum requirements regarding the exposure of workers to the risks arising from optical radiation (EU0412206F). In addition, the European Commission held consultations with the social partners on: the possible revision of the 1994 European Works Councils (EWCs) Directive (EU0405203F); the protection of workers' personal data (EU0211206F); the prevention of musculoskeletal disorders (EU0411204F); and reducing exposure to cancer-causing substances (EU0404202N)

At national level, there was no let-up in the usual flow of new employment-related legislation proposed or adopted in the EU Member States in 2003-4 covering matters such as: part-time work, fixed-term contracts and temporary agency work (eg in Cyprus, Greece, Hungary, Ireland, Malta, Poland and Slovenia); equality, non-discrimination, and reconciliation of work and family life (eg in Cyprus, the Czech Republic, Denmark, Estonia, Finland, Hungary, Ireland, Luxembourg, Malta, Poland and Slovakia); working time (eg in Austria, Cyprus, Germany, Greece, Italy, Luxembourg, the Netherlands and Slovakia); employment, labour market and job creation (eg in Germany, Italy and Spain); health and safety (eg in Estonia, Finland, Greece and Latvia); social security (eg in Austria, Germany, Greece, Italy, Poland, Portugal and Spain); employee involvement (eg in Lithuania, the Netherlands and the UK); termination of contract (eg in Germany, Poland and Slovakia); and general industrial relations issues (eg in Estonia, France, Ireland and the UK). In areas such as atypical work, equality/non-discrimination, health and safety and employee involvement, much national legislative activity was driven by the requirement to comply with EU Directives. In particular, much employment law activity in the new Member States over recent times has involved harmonising their provisions with EU norms.

Trade unions

The estimated trade union organisation rate in Japan fell from 20.2% in 2002 to 19.6% in 2003, continuing a downward trend that has been in evidence for 28 consecutive years. In 2004, US union density slipped to 12.5% of wage and salary workers in 2004, down from 12.9% in 2003 and 13.2% in 2002, with private sector density falling to 7.9% in 2004 from 8.2% in 2003 (in 1983, union density stood at 20.1%).

Few EU countries have union density levels as low as those of Japan or the USA - only France and Spain in the 'old' EU and Estonia, Lithuania, Poland and Hungary among the new Member States. Although comprehensive data are not available, the average density in the expanded EU is probably around 40%, or some 30% when weighted for country size (figures only relatively slightly reduced by EU enlargement), and density is 50% or over in Luxembourg, Malta, Belgium, Cyprus, Finland, Sweden and Denmark. Furthermore, the consistent membership decline witnessed in Japan and the USA, while experienced across the EU as a whole, has not applied to all EU countries. A recent EIRO study of trade union membership over 1993-2003 in 23 EU Member States (excluding the Czech Republic and Lithuania), two candidate countries (Bulgaria and Romania) and Norway (TN0403105U) found that unions have experienced mixed fortunes. Of 19 countries for which total national membership figures were available for the 1993-2003 period, 11 recorded an overall increase in membership and eight an overall decrease (these, of course, are not the same as increases or decreases in union density). The largest percentage increases were recorded in Luxembourg, Ireland, Malta, Norway and Cyprus. The largest percentage decreases were recorded in Bulgaria, Poland and Slovakia. Although more countries registered increases than decreases, aggregate union membership across the 19 countries fell by around a sixth over 1993-2003. There are several reasons for this. The countries where union membership rose were mostly smaller countries, and the absolute increase in union members was thus relatively low. By contrast, the loss of members across the countries experiencing decreases was high in absolute terms because of: massive membership losses in some central and eastern European countries; and significant losses in large countries with high absolute numbers of union members, such as Germany and the UK.

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). In 2004, the US unionisation rate was 13.8% for men and 11.1% for women - though the gap has narrowed considerably since 1983, when the rate for men was 10 percentage points higher than that for women. This narrowing has occurred because men's union membership rate for men has declined more rapidly than women's. The abovementioned EIRO study of European Union membership over 1993-2003, found that, among the union organisations for which data was available in 2003, women made up approximately 41.5% of all members and men around 58.5%. The average share of women in unions in the 'old' EU, plus Norway, was slightly greater, while it was slightly lower (by around one percentage point) in the new Member States and candidate countries. Women made up the majority of union members in Estonia, Latvia, Sweden and Norway; there was near gender parity in Hungary and Denmark; and men comprised two-thirds or more of all union members in Austria, Germany, Malta and the Netherlands. Over 1993-2003, women made up an increasing proportion of union members in most organisations and countries (eg in Austria, Cyprus, Denmark, Finland, Germany, Luxembourg, Norway and Sweden), though with exceptions among central and eastern European countries. Women’s generally increasing share of union membership would suggest that their membership is holding up better than men’s in union organisations that are decreasing in size, and that they are making up a greater share of growth in organisations that are expanding.

One widespread response of trade unions to falling membership (and to other factors, such as changes in the economic structure) has been to merge into larger organisations. In the EU, the trend towards concentration of unions gives the impression of having speeded up over recent years. Over 2003-4, merger activity was reported (usually at the level of the unions affiliated to the main confederations/centres) from countries such as Cyprus, the Czech Republic, Denmark, Finland, Ireland, Luxembourg, Sweden and the UK. In Japan, a move towards merging industry-level union federations affiliated to Rengo started in 2001 and led in 2002 to the creation of the largest federation in the private sector, UI Zensen. In manufacturing, three Rengo affiliates merged to form the Japan Federation of Basic Industry Workers' Unions in 2003. Mergers among transport sector unions are due in 2005. In the USA, there has been some consolidation among member unions of the American Federation of Labor and Congress of Industrial Organisations (AFL-CIO), but the process has been slow. In 2002, several prominent union leaders formed the New Unity Partnership (NUP), which proposes consolidating AFL-CIO's 66 affiliates into 20 or so larger unions, which would organise in specific sectors, industries and labour markets. The debate heated up in 2004, with the results of the 2004 presidential and congressional elections seemingly strengthening the position of the NUP. However, strong opposition to the reform plans among some unions has raised the possibility of a split in AFL-CIO.

Another approach to tackling membership decline is for unions to seek to organise new groups outside their traditional constituencies (key factors in falling unionisation include the loss of employment in traditionally high-unionisation manufacturing industry and the growth of lower-unionisation services employment, and increasing levels of 'atypical' employment). This has been particularly noticeable in Japan over the past few years. The new UI Zensen federation, formed by merger in 2002 and with affiliated unions in the distribution, retail and textile industries, actively seeks to attract part-timer workers, a traditionally low-unionisation group. One in five of UI Zensen’s members is now a part-timer. UI Zensen has also established an occupation-based trade union for care workers, and launched a new union for temporary agency workers (with 20,000 members) in 2004. More part-timers, agency workers and other 'non-regular' workers who were excluded from Japan's traditional enterprise trade unions are joining 'community unions' on an individual basis, as are some foreign workers (see below under 'Migration and industrial relations'). The Japan Community Union Federation (with 5,000 members) joined Rengo in 2003. In the USA, one union bucking the downward trend is the Service Employees International Union (SEIU), which added 700,000 members over 2000-4, making it AFL-CIO's largest union, with 1.6 million members. Its strategies have included organising migrant workers and merging workers' rights with community issues in order to unionise tens of thousands of office-building cleaners in major cities. Unions in many EU countries have also sought increasingly to organise atypical workers (eg in Italy), migrants (eg in Germany) and other 'new' groups, with varying degrees of success, or in some cases to adopt a 'community union' approach (eg several UK unions).

Employer organisations

The 2003-4 period saw a number of developments in employer organisations in the EU. As in recent years, one trend was towards mergers, with the most notable being the fusion of Finland's two main central employer confederations to form Finnish Industries (Elinkeinoelämän keskusliitto, EK), representing at least 15,000 companies that together account for over 70% of Finland’s GNP (FI0403201N). Another example was that in Sweden the two main employer organisations in the municipality and city council sector - the Swedish Association of Local Authorities (Svenska kommunförbundet) and the Federation of Swedish County Councils (Landstingsförbundet) - decided in May to merge by 2007. Short of merger, greater cooperation and collaboration was agreed between central employers' bodies in some cases: the Federation of Finnish Enterprises (Suomen Yrittäjät, SY), a pressure group representing 88,000 SMEs, and Finnish Industries (EK) agreed to increase their cooperation in matters related to labour market policy, giving SY wider influence over collective bargaining (FI0410202N); and the Lithuanian Confederation of Industrialists (Lietuvos pramoninku konfederacija, LPK), which traditionally organises big business, and the Lithuanian Business Employers' Confederation (Lietuvos verslo darbdavių konfederacija, LVDK), which represents small businesses, signed a memorandum and agreement to foster greater cooperation (LT0410102F).

Moving in the opposite direction, 2004 saw a split in the Federation of Employers' Associations of the Slovak Republic (Asociácia zamestnávatelských zväzov a združení Slovenskej republiky, AZZZ SR) formerly the country's only central employer organisation. Dissatisfied members established a new organisation, the National Union of Employers of the Slovak Republic (Republiková únia zamestnávateľov Slovenskej republiky, RUZ SR) which soon grew to around the same size as AZZZ SR (SK0408102F).

A final development witnessed in several EU countries is a reduction in the scale and role of some central employer organisations. It was decided in 2004 that the Confederation of Danish Employers (Dansk Arbejdsiverforening, DA) is to lose one-third of its total budget and close its regional network, transferring its regional activities to member organisations - apparently reflecting a view in its largest affiliate that such a strong central body is not longer required (DK0501103N). In Austria, the Chamber of the Economy (Wirtschaftskammer) is current undergoing an internal reform, aiming to consolidate its budget in the light of a reduction of membership contributions of 30% by 2005.

The most recent significant change in the organisation of employers in Japan came in 2002, when separate central organisations representing companies' employer and trade/business interests merged to form the Japan Business Federation (Nippon Keidanren). The USA is unlike Japan and all EU Member States in that it has no identifiable national intersectoral employers' body with an industrial relations role. Major business organisations such as the National Association of Manufacturers 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 employers' bodies with any bargaining role.

Industrial action

In Japan, the number of collective labour disputes has fallen progressively since the 1970s, to 4,376 in 1980, 2,071 in 1990 and 884 in 2001, though there was a slight upturn to 1,002 in 2002. The main issues at dispute in 2002 were wage increases, followed by dismissal/reinstatement. In the US, industrial action in 2003 was at a very low level by historical standards, with just 14 major work stoppages beginning during the year, 'idling' 129,200 workers and resulting in 4.1 million working days of idleness (about one in every 10,000 available working days). Five work stoppages beginning in 2003 accounted for 82% of all workers idled, and the largest - in supermarkets in southern California - resulted in 67,300 workers idled and 3.3 million days of idleness. This 145-day strike, focusing largely on health insurance issues, finally ended in February 2004 in defeat for the union involved, the United Food and Commercial Workers (UFCW).

Comprehensive figures are not available for industrial action in the EU over 2003-4. Where statistics are available, the picture is mixed. In 2003, strike activity in Austria reached its highest level since the Second World War (AT0411202F) - this resulted from large-scale trade union mobilisation, mainly in opposition to government reform plans related to public pensions and the structure of the state-owned Austrian Railways. By contrast, in Denmark the number of working days lost due to industrial action in 2003 (DK0406101N) was down over three-quarters on the 2002 figure - indeed fewer working days were lost in 2003 than in any year since the relevant statistical series started in 1973, with the exception of 1989.

Austria was not the only country to witness national mobilisations such as short-term general strikes against government policies (notably on pensions - see below under 'Pensions and industrial relations') over 2003-4 - other cases included France (FR0306104F), Greece (GR0412101N), Italy (IT0311102N and IT0412103N), Slovakia (SK0312102N) and Spain (ES0304203N). In individual companies and sectors (public and private), restructuring and attendant job losses gave rise to a considerable amount of industrial action in many countries (notably in central and eastern Europe). The public sector saw high-profile industrial action in countries such as Greece (GR0312101N and GR0411106F), Estonia (EE0312103F), Sweden (SE0306103F), Italy (IT0306102N) and the UK (UK0411105F). Strikes and industrial action within the context of collective bargaining was not prominent over the period, and overall it appears that regular bargaining rounds passed off relatively quietly in most countries, though there were exceptions. For example, in Germany, trade union attempts to conclude an agreement on a 35-hour week in the eastern metalworking industry resulted in four weeks of strike action, but ended in failure. Other examples of conflict over new agreements included action by Swedish electricians (SE0308101N), Slovenian journalists (SI0410302F) and Italian transport workers (IT0312204F) and metalworkers (IT0312102N).

Table 3 below sets out comparative figures on industrial action levels in western Europe, Japan and the USA over 1993-2002. Overall, it seems that, over the decade, Japan had very much lower levels of labour disputes than the USA, which in turn experienced levels around a third lower than the EU (the OECD average was somewhere between the US and EU figures). The trend was downward in the EU and Japan, but slightly upward in the USA. Within the (old) EU, there were very notable differences between a few countries where an average of over 100 working days were lost per 1,000 employees per year - Denmark, Finland, Italy and Spain - and those where 30 or fewer days were lost - Austria, Germany, the Netherlands, Portugal, Sweden and the UK. The new EU Member States are not included in the figures, but is appears that in most cases recent years have seen relatively little industrial action, against a backdrop of economic hardship, high unemployment and so on. However, 2003-4 saw some indications of an increasing propensity among workers to take strike action in countries such as Estonia, Poland and Slovenia.

Finally, in a number of EU countries, new mechanisms for resolving industrial disputes, or to guarantee minimum provision of essential public services, were introduced in 2003-4: examples are Cyprus (CY0404103F), France (FR0412101N), Ireland (IE0408203F) and Italy (IT0310101N).

Table 3. Working days not worked due to labour disputes, per 1,000 employees in all industries and services, 1993-2002
. Annual average 1993-7 Annual average 1998-2002 Annual average 1993-2002 % change, 1993-7 to 1998-2002
Austria 2 1 1 -50
Belgium 27 - - -
Denmark 49 299 177 510
Finland 174 54 110 -69
France 98 - - -
Germany 8 3 5 -63
Ireland 82 73 77 -11
Italy 152 110 131 -28
Luxembourg 12 - - -
Netherlands 27 12 19 -56
Portugal 23 20 22 -13
Spain 283 221 248 -22
Sweden 54 5 29 -91
UK 25 23 24 -8
EU average* 70 59 64 -16
Japan 2 1 1 -50
USA 42 47 45 12
OECD average** 56 49 52 -13

* Excludes Greece and, in some cases, Belgium, France and Luxembourg. ** Excludes, in some cases, Belgium, France, Luxembourg and Turkey.

Source: Labour Market Trends, April 2004, UK Office for National Statistics (based on ILO and OECD data).

Migration and industrial relations

Throughout the world, international migration has increased substantially in recent years. The United Nations estimates that the number of international migrants worldwide stood at 175 million in 2000, up from 76 million in 1960. Furthermore, during the 1990s, international migration to developed countries increased significantly, even though many of them restricted the admission of migrants. These developments are closely linked to the labour market. According to the 2004 edition of the OECD’s annual Trends in international migration, record numbers of people are moving to many OECD countries in search of jobs and to join their families. It notes that the economic downturn in some OECD countries has not affected the upward trend in international migration that began in the mid-1990s. There was a significant increase in labour-related migration during 2001-2, of both temporary and permanent workers and across all employment categories - skilled workers, seasonal employees, trainees, 'working holiday-makers,' transfers of staff within multinational companies, and cross-border workers. Other categories of admissions such as family unification, refugees and students also continued to grow.

In this context, the impact of labour migration on industrial relations, and how industrial relations systems deal with migration, are increasingly relevant questions. Migration can play a role in meeting employers' needs for labour, in the light of demographic change or skills shortages, while at the same time raising questions such as the relationship between migrant workers and nationals who are unemployed, or the fair treatment of migrant workers in employment, both on grounds of equity and related to concerns about 'social dumping' in the form of a pool of migrant workers employed on a low level pay and conditions, or willing to accept employment on such terms. In this context, it may be expected that migration will have major effects on industrial relations systems and become an issue for the social partners and in dialogue and bargaining between them (as well as a theme in government employment policy and legislation). Some aspects are explored below (the EU information is largely based on a 2003 EIRO comparative study, which covered the old EU and four new Member States, TN0303105S) - while box 4 sets out some basic statistics on migration in the EU, Japan and the USA (it should be noted that the categories and terminology used in the area of migration vary from country to country and from source to source).

Box 4. Migration figures The number of foreign nationals in Japan has increased steadily since the mid-1980s. The number of registered foreign nationals rose from some 900,000 in 1987 to 1.3 million in 1992, and the number of foreign nationals stood at 1,915,000 in 2003. The total number of foreign workers stood at 580,000 in 1992, or 1.1% of the workforce, rising to 710,000 (1.3%) in 2000, 740,000 (1.4%) in 2001 and 760,000 (1.4%) in 2002. Migration to the USA has risen sharply in recent years, with more immigrants arriving during the 1990s than in any previous decade. The number of legal migrants averaged 935,000 per year from 1991-2002. In 2001 and 2002, some 1,064,000 were admitted in each of the two years, but the number fell to 705,000 in 2003 as a result of new checks and restrictions introduced after the events of 11 September 2001. Immigrants presently make up 11.3% of the US population. They constituted 14% of the US labour force in 2003, more than double the 1980 percentage. There are some 18 million foreign-born workers in the labour force, plus 5.8 million people admitted on a short-term basis for business purposes and 990,000 'temporary foreign workers'. In 2003 (according to the European Commission), of the 381 million people living in the EU15 (no similar data are yet available for the expanded EU), some six million (1.6%) were EU citizens living in other countries and 14 million (3.7%) were non-EU nationals, while the total number of people not residing in their country of birth or citizenship was estimated at 33 million-36 million (8.7%-9.4%). The EU15's migrant population represented roughly 8% of the total population. There is substantial variation between countries with Austria, Germany and Belgium having the highest share of people with a foreign nationality, at 9% or more. The share of foreigners is low, at 2.5% or less in Greece, Finland, Portugal, Spain and Italy. No overall figure is available for the proportion of the total EU workforce made up by non-nationals, but the European Commission reports that in most Member States, the share of non-nationals is somewhat lower in the labour force than in the population as a whole. This is the case in Belgium, Germany and Austria, where it lies between 8% and 10%, in the Netherlands and Sweden, where it is about 4%, and in Italy, Finland, Spain and Denmark, where non-nationals account for 1% or 2% of the labour force. On the other hand, non-nationals are at the least equally as represented in the labour force as in the total population in France (6%), Ireland, Greece and the UK (3% to 4% ) and Portugal (2%). The overall picture is thus that migrant workers make up a much larger proportion of the labour force in the USA than in Japan, in fact around 10 times larger. The EU15 as a whole lies around half-way between the two extremes, with some countries approaching US levels (eg Austria, Germany and Belgium) and some near Japanese levels (eg Finland, Greece and Portugal).

Labour market situation

Key aspects of the labour market and employment situation of migrant workers include the following.


In the EU as a whole, foreign workers enter the full spectrum of occupations, but seem to be concentrated largely at the top and bottom of the labour market. Many migrant workers are in professional or skilled occupations - according to Eurostat labour force survey figures, 23.3% of non-EU national workers were in high-skilled non-manual occupations in 2002. However, the largest group of migrants is found in unskilled manual work, accounting for 33.7% of non-EU national workers in 2002. A similar picture is reported from the USA, with immigrants having a disproportionate presence in both high-level occupations - such as computer scientists (18.6% of whom are immigrants) and engineers (16.4%) - and low-skilled ones - immigrants make up 14% of the labour force, but 20% of the lower-paid workers. While no statistics are available, in Japan the thrust of official policy has been to admit highly qualified professional/technical foreign workers, while discouraging an influx of unskilled workers, suggesting that migrants (legal ones at least) are found more in the upper strata of the labour market (though see next point).


The sectors where migrant workers are employed tend, across the EU, to be generally labour-intensive and lower-skilled. In 2002, the main sectors of employment for non-EU nationals were manufacturing (22.0%), wholesale/retail trade (12.9%), business services (11.9%), hotels/restaurants (11.2%) and construction (10.6%), according to Eurostat. The proportion of non-EU nationals employed in hotels/restaurants and private households is far above that of EU nationals and, to a smaller extent, this is also true for manufacturing, construction and business services. In the USA, it appears that key sectors for migrant employment are manufacturing (eg textiles), construction, agri-business and homecare services. In Japan, manufacturing is very much the dominant employment sector - a 2003 survey found that manufacturing industry hires 60.2% of foreign workers, followed by services (9.0%) and the restaurant/hotels (7.4%).

Employment status

Migrants in the EU Member States are more likely than average to have 'flexible' and fixed-term contracts. To take the latter example, across the EU these were held by over 20% of non-EU nationals in 2001, compared with 13% of EU nationals (Eurostat figures). According to a 2003 survey in Japan, only 25.8% of foreign workers worked on a full-time regular basis. While no statistics are available for the USA, large number of migrants are admitted only temporarily for employment purposes, while many work for temporary work agencies or on a day-labour basis.


The relatively unfavourable situation of many (though not all) migrants on the EU labour market is reflected in their wage level (and illegal immigrant workers are likely to be in an even worse position). Migrants tend to be among the low-paid groups in a number of countries (TN0208101S). Detailed information on this point is rarely available, but widespread low pay is reported from countries such as Spain and Belgium. As noted above, immigrants make up 14% of the US labour force, but 20% of the lower-paid workers, while half of the immigrant workforce earns less than 20% of the minimum wage, compared with one-third of native-born workers. No similar data are available for Japan, but the low-pay effect may be less pronounced, given that legal immigrants appear to be found at somewhat higher levels of the labour market.

Working conditions

Overall, it appears that migrant workers in the EU are often worse off than nationals where working conditions are concerned (with illegal immigrants worst off of all). The problems or practices faced to a greater extent by foreign workers than nationals may include: working on weekends, holidays, at night or in alternating shifts (as in Germany); less job security (Spain); inadequate training (the UK); excessively long working hours (the UK); and poor health and safety conditions or accident rates (Austria, Spain and the UK). Much of these poor conditions can be explained by the job and sectors in which many migrants are employed. The situation in the USA seems very much the same. There is a large body of evidence indicating that many immigrants are abused by employers, injured as a result of employer negligence or dismissed for trying to form a union. Studies suggest that employer violations are quite widespread. A General Accounting Office report has disclosed that denial of labour rights is almost routine in the 'sweatshops' that have proliferated in cities like New York during the last two decades. Other studies have found that immigrants are disproportionately exposed to violations of occupational health and safety laws as they are less likely to receive necessary safety equipment and training. Studies and anecdotal evidence also point to pervasive wage and hour violations committed by day-labour employers and temporary work agencies. There is also reportedly a lack of employer-provided training programmes available to workers who are poorly educated or lack English proficiency in order to improve their skills and prospects. Criticisms regarding a lack of training opportunities have also been levelled against government workforce development programmes. No information on these issues is available for Japan.

Trade unions and migrants

The fact of labour migration poses a number of questions for trade unions. There is arguably a tension between 'traditional' attitudes of wishing to protect the jobs, pay and conditions of their existing members against competition from newcomers (who may be prepared to work at a lower cost), and the prospect of a potentially fruitful recruiting ground (at a time of generally falling membership) among workers who often experience the kind of poor pay and conditions that prompted the birth of many trade unions in the first place. In the EU, this tension is apparent in a number of countries, with some unions sceptical about the desirability of further labour immigration and opening borders for the purpose of 'importing' further foreign workers - perhaps the clearest example is Austria. Furthermore, unions in countries such as Denmark, the Netherlands, Belgium and Ireland stress the problem of unemployment among migrants already in the country. Less concern about a possible 'threat' to their members from new migration seems to be the case in countries such as Italy and the Netherlands. The EU's eastward enlargement in 2004 brought specific worries about the effects on national labour markets among unions in countries such as Austria, Finland and Germany (especially in the construction sector). Such worries are not restricted to the old Member States - for example, unions in Hungary are concerned about the import of cheap labour from neighbouring eastern European countries. Despite their concerns, however, many unions in the EU act to support or recruit migrant workers - as explored below under 'Activities'.

More restrictive attitudes still appear to dominate in many Japanese unions. For example, the Rengo confederation is in favour of admitting only 'skilful, professional, technical foreign workers' and attaches high value to 'national consensus and proper balance with job opportunities for Japanese workers'. Citing a need to protect Japanese workers and prevent a deterioration in working conditions, trade unions have consistently opposed opening up the labour market to foreign workers, including unskilled labour. Further, because most Japanese trade unions are enterprise unions comprising mainly full-time workers, they see almost no incentive to organise or give support to foreign workers, and many do not admit them. More open policies have recently been developed by some unions, but with limited success (see below under 'Activities')

By contrast, many US unions are now giving priority to the interests of immigrant workers. Historically cool towards immigrants out of fear of labour market competition, many unions now regard an alliance with immigrants around civil and workers' rights issues as a potentially critical dimension of reviving the trade union movement. This has led to a number of major initiatives and organising drives (see below under 'Activities').


Although there is a lack of statistical information for most EU countries, trade union density among migrant workers seems to be lower than the average - as in Belgium, Denmark, Hungary, Ireland, the Netherlands and the UK. However, there are often considerable sectoral variations within countries where the overall density for migrants is below average. For example, in the UK, the unionisation of foreign workers is thought to be lower than the (low) average in hotels/restaurants and wholesale/retail, but nearer the (high) average in public education and health services. Similarly, in Denmark, unionisation of migrant workers appears to be lower than average in sectors such as hotels/catering and cleaning, but nearer average levels elsewhere. Overall, indeed, it seems that an important factor explaining their below-average union density in many Member States is that many migrants work in service sectors where unionisation is traditionally low. However, the picture of notably low relative unionisation among migrant workers is not universal. In Sweden, for example, union density among foreign workers is only marginally lower than the 80% figure for Swedish nationals. Even more notably, in the Emilia Romagna region of Italy, research indicates that union density among migrants is significantly higher (45%) than the average (36%).

There appear to be no overall data on the level of unionisation among migrant workers in the USA. One piece of evidence is that in California (one of the main states of residence for immigrants), in 2001-2, the unionisation rate for immigrants was 11.7%, compared with 19.7% for US-born workers (State of California labor 2003, University of California Institute for Labor and Employment). Similarly, in Illinois research published in 2002 found that 12% of immigrants in employment in Chicago belonged to a union compared with 19% of all workers in Illinois (Chicago's undocumented immigrants: An analysis of wages, working conditions, and economic contributions, Chirag Mehta et al, Center for Urban Economic Development, University of Illinois at Chicago, 2002). However, some unions appear to have recruited considerable numbers of immigrants in recent years, as highlighted below.

Again, no union density data for foreign/migrant workers are available for Japan, but it seems that the level is low. According to a 2003 survey, only 11.3% of establishment-level trade union organisations have foreign workers as members, and a majority of them actually exclude them (see next point).


Seeking to recruit migrant/foreign workers is a very obvious way in which trade unions can attend to their needs and interests (and boost their own often declining memberships). However, in the EU, union recruitment initiatives directed specifically at migrant workers are reported only from a minority of countries, such as Finland, Italy, the Netherlands, Spain, Sweden and the UK, while in some cases, efforts are being made to improve the involvement of migrant members in trade union work - eg Finland - and their representation in union structures and among union officials - eg Belgium, Italy and the Netherlands. The US union movement took an unprecedented turn towards immigrant workers in 2002, identifying them as a source of revival. A leading example is the Service Employees International Union (SEIU) which has grown rapidly to become the largest affiliate of AFL-CIO partly as a result of immigrant organising - for instance, its 'Justice for Janitors' campaign during the 1990s unionised tens of thousands of office-building cleaners, often immigrants, in major cities such as Los Angeles. As noted above, Japan's enterprise-level unions have in most cases made no efforts to recruit foreign workers, while around three-quarters do not even permit membership for such workers and mostly have no intention of making efforts to attracting foreign workers as members. Efforts to organise migrant workers have thus been largely restricted to the very much less prevalent community-based general trade unions ('community unions'). In addition, some industry-level union federations organising small-scale manufacturers where a large number of foreign workers are employed have recently initiated efforts to attract foreign workers, and to encourage the establishment of new trade unions. However, these efforts are facing difficulties because, for example, many foreign workers are not permanent residents and will not, it is thought, be reliable union members, while communication may be difficult due to language differences.

Other possible types of trade union activity related to migrant workers include:

  • involvement in, or the instigation of, campaigns and initiatives to combat racism and promote equal treatment for migrants. Such actions are reported from the majority of western European countries, while specific campaigns to promote equal rights for migrant workers - often in non-employment areas (such as housing and political rights) as well as in employment - feature in many countries. Campaigning around such issues plays a major role in the USA. Notably in 2003, AFL-CIO organised a national mobilisation of immigrant workers, known as the Immigrant Workers Freedom Ride, centred on full citizenship rights for immigrant workers, clear opposition to civil rights abuses and discrimination against immigrants, and enforcement of immigrant workers’ rights in the workplace. No such activity is reported from Japan;
  • practical trade union assistance and support for migrant workers and their labour market and social integration. This is a feature in many 'old' EU Member States, often taking the form of advice, information, special services and involvement in integration programmes. Similar efforts are present in the USA but rare in Japan: though a number of support initiatives have been taken by a number of Japanese community-based and industry-level union organisations (with some concern about the financial burdens), it appears that NGOs rather than trade unions provide support to foreign workers in most cases; and
  • in EU countries with large numbers of illegal immigrants, these seem to be a major issue for trade unions. Thus unions in countries such as Greece, Portugal and Spain devote considerable attention to supporting illegal foreign workers (and recruiting them in some cases) and seeking to regularise their situation and defend their rights. Some US unions also organise illegal immigrant workers, but it is reported from Japan that neither unions nor illegal workers themselves are in favour of their joining unions.

Employers, employer organisations and migrants

Labour shortages, both present and future, seem to be the main reason for employer organisations in western Europe to develop stances and policies on migration and immigrant workers (in the new EU Member States, there is very little in the way of employer organisations' policies or activities in this area). Organisations in countries such as Germany, Belgium, the Netherlands and the Scandinavian countries point to long-term demographic developments, with a declining number of young people and an increasing number of older people. As well as reducing the size of the workforce, this may weaken the basis of the social security and pensions system. Not only at present, but also in the near future the entry into the labour market of migrant workers is widely perceived as necessary by employers in the great majority of countries. With few exceptions (eg Austria and, to a lesser extent, France), the idea of a free market in trade and people, especially within Europe, is widely embraced by employer organisations, which thus press governments to reduce and simplify regulations. However, in most countries, their interest in such a free market in people is conditioned by their own demands for labour and specific shortages. The majority of employer organisations in western Europe are thus of the same opinion on the necessity of immigration to meet labour shortages. In many countries, employers criticise governments for being too slow, bureaucratic and ponderous on these subjects, and have made proposals to the government to tackle these problems.

In Japan, employer associations have hitherto supported the current, relatively restrictive, legal framework governing the entry of foreign workers, but they too anticipate labour shortages in the near future and recognise a need to attract more foreign workers in an acceptable manner, by easing immigration controls in the medium and long term. This view was expressed clearly in 2004 by Nippon Keidanren in a set of 'recommendations on accepting non-Japanese workers', which calls for legal change to facilitate the necessary labour migration. Many individual Japanese employers, however, appear to take a cautious approach to opening up the labour market to foreign workers. According to a 2003 survey, 45% of respondent companies showed positive attitudes toward foreign workers while about 44% had 'no intention of hiring foreign workers'. As to a desirable national policy on foreign workers, almost 40% of respondents supported selectively accepting foreign workers for the job categories that suffer from labour shortages, while some 60% wanted strict controls over foreign workers via work permits based on their qualifications or job experience.

The views of US employers mirror those expressed in many EU countries and, to a lesser extent, Japan. In recent years, they have tended to favour a generally 'open-door' approach to immigration and have resisted political efforts to reduce the inflow of new immigrants. This is true for employers that tend to hire technical and professional workers, as well as those that rely on low-wage unskilled labour. The National Association of Manufacturers, for instance, sees immigration as one way of attracting sufficient numbers of sufficiently trained workers, given a lack of suitable 'home-grown' workers and the coming retirement of the 'baby boom' generation. Employers's demands for a steady supply of highly-skilled technical and professional workers rest on the argument that the world’s 'best and brightest' should be allowed to come to the USA, as they help US companies survive and prosper in the global marketplace.

In terms of specific activities by employer organisations relating to migrant workers, apart from involvement in various bipartite and tripartite dialogue and initiatives (see below), these are not that common in the EU. Examples include: a declaration on anti-discrimination and equal opportunities in recruitment of immigrants in Belgium; the organisation of surveys and seminars on migration issues in Finland; support for company-level anti-discrimination measures in Germany; action on recruitment of people from ethnic minorities in the Netherlands; activities in some sectors to coordinate labour migration and combat illegal immigration in Portugal; and projects and policies on the integration of migrants, and cooperation with local authorities on placement of migrants in Sweden. A notable example is Italy, where some employer organisations (and individual employers) have taken various initiatives - such as training to upgrade skills and Italian-language courses, or help with housing - aimed at assisting the integration of migrant workers in the workforce, and in society more widely, taking the view that social inclusion may help to improve the reliability of workers and the stability of the employment relationship. At the level of individual employers, in some countries there are companies that support migrant workers by addressing discrimination and operating 'diversity management' practices - eg in Denmark, Ireland and the UK. Turning to Japan and the USA, there seems to be little in the way of explicit activity relating to migrant workers emanating from employers or their organisations (beyond recruiting them and expressing views on public policy). Japanese research has indicated a recognition among employers of a need to raise awareness of the issue among employees, but there is no evidence of concrete action on this point.

Consultation and bargaining

Reflecting national systems of social partnership, trade unions and employer associations in most EU countries (exceptions include France and Hungary) are consulted by governments on issues relating to migrant workers. The level of consultation varies, however, from consultation on specific issues such as quotas (eg in Austria, Slovenia and Spain) to involvement in the drafting of all relevant policies. To take a few examples: in 2002 the Danish social partners, municipalities and government concluded an agreement on stronger measures to integrate immigrants and refugees into the labour market; in Germany, the social partners were involved in the 'immigration commission' that drew up a report which formed the basis for new immigration legislation in 2004; the Italian social partners are consulted in the drafting of migration policies and are represented on national and local bodies dealing with immigration issues; and the social partners are represented on a recently established working group on illegal immigration in the UK.

In Japan, where consensus-building is also a feature of much policy-making, some social partner involvement is reported in the migration area. Notably, in the context of current proposals to ease restrictions on labour immigration, government ministries are holding discussions with representatives of trade unions and employers in order to determine overall government policies and specific approaches for: carrying out proper immigration control; maintaining public order; protecting the human rights of foreign workers; and examining possible increases in training/education or social security costs, as well as who will meet them. Given its absence of a tradition of social partnership or tripartite concertation, it is unsurprising that no explicit involvement of unions and business/employer organisations in migration issues is reported from the USA.

Across the EU, the issue of migrant workers does not appear to have achieved a significant place on the collective bargaining agenda. In many countries, issues relating to migrant workers are not seen as an issue for collective bargaining, but for bipartite and tripartite dialogue. There is no evidence of any bargaining activity of specific relevance to the matter in France, Luxembourg, Portugal, Sweden and the new Member States. In the other countries, there is some - though rarely very much - bargaining activity on migrant worker-related themes. In Austria, this is purely negative, in that a number of sectoral agreements allow discrimination against migrant workers in recruitment and dismissal. Elsewhere, the approach varies, with agreements at various levels dealing in various ways with a number of key themes, mainly non-discrimination/equal treatment on grounds of race and ethnic origin (which is relevant to many migrant workers, as well of course to many nationals) and integration. At intersectoral level, the Belgian social partners have reached agreements both on equality and non-discrimination on racial grounds, and on special training and employment measures for 'at-risk' groups, including migrants. Equal treatment for migrants is established as a theme in company-level employee-management cooperation by a national agreement in Denmark, while the largest social partner organisations have also reached an agreement on the integration of migrants. Non-discrimination and equal treatment is also dealt with by intersectoral agreements in Greece and Ireland. The narrower issue of monitoring the working conditions of migrant workers has been addressed by a Finnish central agreement. In most cases, there is no evidence that such intersectoral provisions are reflected in any significant bargaining on migration issues at lower levels.

The countries where migrant workers probably receive most attention in collective bargaining are probably those where the issue is dealt with in sectoral agreements - Italy, the Netherlands and Spain. Around one in 12 Italian sectoral agreements deal with relevant matters, as do around a fifth of major Dutch agreements (the majority of which are at sector level) and a smaller number in Spain. In Italy, the relevant agreements - often signed by employer organisations for SMEs, crafts and cooperatives - cover a range of matters including: support for employment and social inclusion (housing, transport etc); language courses and, less frequently, vocational training; the creation of special monitoring committees on migrant workers' labour market situation; equal opportunity measures; and special longer holidays for migrant workers (ie to visit their countries of origin) and time off for religious purposes. In the Netherlands, the focus is more specifically on the employment of migrant workers, with a majority of relevant agreements containing target figures for new jobs for this group. In Spain, collective agreements in sectors with large migrant workforces - eg hotels/catering, construction and agriculture - include provisions on issues such as equal pay and (as with some Italian agreements) flexible working time for religious purposes, such as Ramadan for Muslims.

Company-level bargaining on migration issues seems to be rare, with some exceptions in Germany, Italy, the Netherlands and possibly the UK. A few company agreements in Italy provide for special working time and leave arrangements for migrant workers. In Germany, works agreements on preventing racial discrimination and ensuring equal treatment have been signed by works councils and management in a number of companies (covering around one million employees).

If bargaining on topics directly related to migrant workers is very patchy at best across the EU, it appears to be non-existent in Japan. A lack of data on the content of collective agreements makes the situation difficult to assess in the USA, but many unions are now actively championing immigrant workers' issues, so it seems likely that some relevant bargaining activity is taking place. Immigration is also very much an industrial relations 'issue' in the USA in ways that would arguably be very unusual in the EU. There are reports of immigrants being dismissed for trying to form a union in low-wage sweatshop sectors. Notably, some employers have been found to be using immigration law to break unionisation efforts. A number of incidents have occurred where employers have informed on workers seeking to form a union or being involved in union activity, by contacting the authorities and inviting them to take action against suspected illegal immigrants. The requirement for workers to have a valid social security number is, it is claimed, used by some employers as a means of threatening suspected illegal immigrants not to protest at poor pay or working conditions that may themselves be illegal.

Pensions and industrial relations

Pensions systems and their reform are currently high-profile issues in the EU, Japan and the USA, as in most industrialised market economies. State pension schemes have recently been reformed in many EU Member States and in Japan, with further changes to come in many cases, while the US system is due to be overhauled soon. In numerous EU countries, governments have raised the pension age and/or are developing plans to do so, and early retirement schemes are under heavy pressure. Other common changes include higher contributions, lengthened contribution-history requirements, and basing pension entitlements on a longer period of earnings. The main reason usually cited is that demographic change, resulting in an ageing overall population, threatens to make the costs of the present systems prohibitive, and that measures must be taken to make them sustainable (as well as to keep people in work longer). Examples of EU countries that have recently introduced major state pension reforms include Austria, Belgium, Finland, France, Germany, Greece, Italy, Portugal, Slovakia and Spain. Against a similar background of population ageing, in Japan a process of gradually increasing the pensionable age from 60 to 65 started in 2001. In 2004, a new pension reform plan provided for a gradual increase in contributions, new benefit calculation formulas and changes in financing arrangements. The US 'social security' pension system has not been reformed since 1983, when the retirement age was raised to 66 (starting in 2005) from 65. However, again in a context of demographic change and (disputed) concerns about long-term sustainability, the Bush administration re-elected in November 2004 has proposed the creation of private tax-free investment accounts within the social security pension scheme: workers could divert some of their social security contributions into personal investment accounts in exchange for agreeing in advance to receive a much-reduced guaranteed state benefit when they retire.

The reform of state pension provision has increased attention on employment-related occupational pensions, and in many countries the regulatory environment has been changed in recent years in order to encourage such 'second-pillar' provision. This is true in many EU Member States - though the relative importance of occupational pensions varies enormously among them, as do the characteristics of schemes - as well as in Japan and the USA. Table 4 below highlights some key aspects of occupational pensions systems in the various countries.

Table 4. Main features of occupational pensions systems, EU, Japan and the USA
General Among EU countries, collective (and often sector-level) occupational pension schemes are deeply rooted in the Netherlands, Sweden, Denmark and, to a lesser extent, the UK. Relatively recent legislative reforms have encouraged the spread of such schemes in countries such as Austria, Belgium, Germany and Greece. Other types of occupational pension provision include pensions provided through insurance firms (which exist, for example, in Austria, Finland and the Netherlands) and 'direct pension guarantees' made by the employer (eg in Austria, Belgium and Finland), while distinctive country-specific schemes are present in France and new Member States such as Hungary and Poland. In Japan, a system of occupational pensions has existed in the private sector since the 1960s. Against the background of increasing mobility and diversification of employment, the system was reformed in 2002 in an attempt to protect pensioners, secure pension benefit portability and generally improve the environment for occupational pensions. The new schemes may be either corporate or individual type in nature. Occupational pension programmes are usually instituted by a company or group of companies, and fund management duties are entrusted to pension management institutions (private insurance companies, banks etc). Occupational pensions have a relatively long history in the USA. Traditional defined-benefit schemes are generally collectively based, covering the workforces of individual employers or, in a significant proportion of cases, a number of employers. The newer and growing defined-contribution schemes (see below) are based on individual employees. Congress has taken a number of measures to facilitate the expansion of occupational pension coverage, by offering various incentives for companies to establish plans.
Coverage In the EU, it appears that some countries are approaching full workforce coverage by occupational pensions, for example in Denmark, the Netherlands and Sweden, as well as in France, if this country's mandatory supplementary schemes are treated as occupational pensions. In Germany, Norway and the UK (plus Hungary, if its second-pillar funds are counted as occupational), coverage is around half of the workforce, or a little more, while a third to a half appears to be the rate in Belgium and Ireland. Only around a sixth of Austrian workers are covered by a scheme. In Finland, Greece, Italy and Spain, coverage appears to be below 10%, in some cases far below. Overall coverage figures are not available for Japan, but according to a 2003 survey of companies with 1,000 or more workers, 90% operated occupational pension schemes, plus retirement lump-sum grants. However, the number of employees covered by traditional occupational schemes decreased by three million in 2003, continuing a decline from peak participation during the mid-1990s. In 2003 the percentage of US workers participating in an employment-based retirement plan was 42% (57.1% for full-time, full-year workers aged 21-64) - 63.5 million workers in total (down from 67.1 million in 2000).
Variations in coverage Apart from differences in overall occupational pensions coverage, there are large variations across different types of organisations and types of employees in the EU. For example, overall it appears that: coverage in larger enterprises is usually higher than in SMEs; coverage usually varies between sectors, and coverage in the public sector is often higher than in the private sector; and the likelihood of employees being covered by an occupational pension varies with factors such as their sex, job level and type, employment status and length of service. On the last point, in many countries, coverage tends to be higher among white-collar workers than blue-collar workers, men are better represented than women and coverage of 'atypical' employees (part-time, fixed-term etc) is below average. Efforts to tackle these discrepancies are being made in some countries. A lower prevalence of occupational pension schemes in SMEs also appears to be the case in Japan. This is clearly true in the USA, though coverage levels have increased in smaller firms (eg, in companies with 25-99 employees, the level rose from 28.2% to 39.4% between 1987 and 2003). In the past two decades, a large coverage gap has opened up between public and private sector full-time workers in the USA. Most public sector workers (98% in 1998) participate in retirement plans run by their respective state legislatures, but the fall in the level of private sector coverage is striking: in 1985, 91% of private sector full-time employees in medium and large establishments participated in a retirement plan, but by 2000 the level had fallen to 70%. Coverage is also considerably lower among less well-educated workers in the USA.
Defined-benefit v defined-contribution The 'traditional' form of occupational pension in the EU is the defined-benefit (DB) scheme, whereby the pension and other benefits that will be paid to members are clearly defined in the scheme’s rules. The other main form is the defined-contribution (DC) scheme, whereby the member’s benefit is determined solely by reference to the contributions paid in by the employer (and, if contributory, by the member) and the investment returns earned on those contributions. DC schemes have long been the main form in Austria, Denmark and Italy, while a shift from DB to DC schemes is reported from a relatively large number of countries - this is a particularly major issue in Ireland and the UK. Furthermore, countries that have only recently set up occupational pension schemes have mostly done so in the form of DC schemes, as in Poland and Hungary. However, DB schemes are still dominant in countries such as France, the Netherlands, Finland (though here occupational pensions are of minor importance) and Germany. In cases such as Austria, Belgium and Sweden, DB and DC systems exist alongside each other, without any shift being reported from the former to the latter. In Japan, many companies have been dissolving their traditional, non-DC occupational pension funds, given poor investment yields and new accounting standards. Furthermore, the traditional pension schemes are not seen as being compatible with the current lower birth rate, frequent job switching and diversified work patterns. Legislation adopted in 2001 will eventually abolish the old schemes, replacing them with new DB or DC schemes. It is expected that more and more employers will introduce DC plans or another new scheme, and in 2003 some 700,000 employees joined a DC scheme. In the USA, over the past two decades there has been a decisive shift away from DB schemes to DC plans, with a declining stock market, low interest rates, and a falling active worker-retiree ratio making the former less attractive to employers. DB schemes have suffered a sharp decline over the last 20 years, as measured by both the number of plans that have been terminated and the number of workers covered. As the same time, DC plans have increased dramatically - especially in the private sector - and now cover 59 million workers, compared with 44 million for DB schemes.
Final salary v average salary Apart from a relatively widespread shift from DB towards DC schemes, in some EU countries a shift is occurring within DB schemes, from using the employee's final salary at retirement to calculate the amount of the pension to using their average salary over a varying period of employment - examples include the Netherlands, Ireland and the UK. Average-wage schemes have a longer history in cases such as Finland and the French private sector. The main motive of the move within DB schemes towards average-salary arrangements is financial, with employers complaining that the costs of DB pensions are becoming excessive. In the USA, within the declining DB sector, benefits are still usually based on a combination of the workers’ age, years of service and final average earnings.
Contributions Contributions to occupational pension schemes are shared in varying measures by employer and employees in many EU Member States, with the employer invariably paying the majority in such cases. For instance, employers generally pay around two-thirds and employees around one-third in Denmark, the Netherlands, the UK (DB schemes) and France (60%/40%). The share paid by employers may be much higher (or even 100%) in some countries, as in Spain, Poland, Finland, Austria and Sweden. In Japan, employers alone pay the contributions for the new DB and DC pension plans, and under one of the traditional schemes (the 'tax-qualified pension plan'), while employers and employees basically bear equal payment burdens for the other traditional scheme (the 'employees pension fund'). In the USA, both employers and employees contribute to DB schemes. In DC plans, employees accumulate money for retirement by making pre-tax contributions from their salary. Employers often make a limited contribution to the plan, but monthly contributions are not required.

Source: EIRO.

In this section, we examine the differing relationship between pensions systems and industrial relations in the EU, Japan and the USA (the EU information is largely based on a 2004 EIRO comparative study, TN0404101S, covering the old EU and two new Member States). The main focus is on occupational pensions, as this is the area of pensions where, in many countries, the social partners, social dialogue and collective bargaining play the greatest role. First, however, we look at the involvement of the social partners in the state pensions system.

State pension systems - involvement of trade unions and employer organisations

In the EU, the formal position of the social partners in managing the 'first pillar' of state pension provision varies from very strong, as in France, to virtually none, as in the UK. Often, the social partners have an advisory or consultative status - eg in the Netherlands or Hungary. Although in several countries the social partners have little or no formal influence, their role in practice might be considerable - examples include Finland, Belgium and Denmark. In most EU countries concerned, recent state pensions reforms have involved consultation with the social partners. In Spain, for example, the social partners are involved in negotiations over all aspects of the public pension system, with recent changes based on agreements with all or most social partners. The social partners are not, however, always unanimous about pension reform in countries where they are involved in the process. In Italy, a 1995 reform law (IT0309203F) made structural changes to the state pension system on the basis of an agreement between the trade unions and the government, which was not signed by employer organisations. Furthermore, consultation does not always mean that the social partners' views are taken into account: in Austria, for instance, though governments have always consulted them prior to making significant changes in the state pension system, in 2003 the current conservative-populist coalition administration carried through a fundamental pension reform against the explicit opposition of the social partners (AT0305202F).

In the USA, the social security pension scheme is governed by Congress, and trade unions and business/employer organisations are not involved, though they lobby on the issue and over possible reforms. While not apparently involved in managing the state pensions system, the Japanese social partners are consulted on legislative reforms, such as those in 1999 and 2004. Indeed the Rengo trade union confederation and Nippon Keidanren employer confederation issued joint statements expressing their opposition to some aspects of the 2004 reform.

Occupational pensions - collective bargaining

In most 'old' EU countries, there is an appreciable degree of collective bargaining over occupational pensions (Finland and Luxembourg, where this sort of pension provision is relatively uncommon, are exceptions), though such bargaining exists in only a very limited way in new Member States such as Hungary and Poland. The extent of bargaining over occupational pensions in each country depends on the relative importance of such schemes, and the involvement of social partners in them, but there does appear to be a general tendency for bargaining - especially at sector level - to play an increasingly important role in this area, often encouraged to do so by recent legislation promoting occupational pension provision.

National intersectoral bargaining on what can be considered occupational pension-type provision occurs in France and Ireland. Sectoral collective agreements are essentially the basis for the entire system of occupational pensions in Denmark, the Netherlands and Sweden. Such agreed sectoral arrangements obviously result in a high degree of coverage, with up to 90% of the workforce covered by the schemes in these countries. Under the influence of recent legislation, a similar system of sectoral agreements on occupational pensions covering all major branches now appears to be developing in Germany and Italy, though coverage is not yet at such high levels. There are a few examples of relevant sectoral collective agreements in Austria, and the trend appears to be to a greater focus on this level of bargaining over occupational pensions. Belgium has a history of sectoral agreements on occupational pensions in a number of industries, and a legislative reform that came into force in early 2004 should boost bargaining on this theme at sector (and company) level. Recent Greek legislation seeks to do the same. Company schemes are generally based on work agreements between works councils and management in Austria, while company-level bargaining over aspects of occupational pensions occurs in unionised workplaces in Ireland. In Spain, the limited amount of collective bargaining on occupational pensions occurs exclusively at enterprise level. Company-level bargaining over aspects of occupational schemes, especially when they are faced with problems, seems to be spreading in the UK. Although the sector level predominates, there are schemes based on company agreements in Denmark, Italy and the Netherlands. It is arguably in the Netherlands that occupational pensions are the hottest issue in (sectoral) bargaining at present, with the key points at issue being the level of contributions and the nature of the schemes.

Where bargaining on occupational pensions exists, its relationship with 'normal' bargaining over pay and other conditions varies in the EU. In Sweden, for example, the agreements on occupational pensions are separate from (and have different coverage than) ordinary agreements on pay and conditions. There is also little linkage in France, where occupational pensions are negotiated at intersectoral level and pay bargaining occurs at lower levels. In Denmark, Ireland and the Netherlands, however, the issue of occupational pensions is often part of general pay bargaining. In Denmark, possible increases in pension contributions are often part of the overall bargaining trade-off agenda, while in the Netherlands recent rises in contributions, resulting from funds losing a major part of their reserves because of stock market problems, have taken away a large part of the scope for pay rises in bargaining rounds

In Japan's normal collective bargaining process at enterprise level, management of the corporate pension will be on the agenda only when the employer suggests any changes in the system, or the introduction of a system, that may involve a change in employment conditions. In many cases collective agreements stipulate retirement benefits. This means that before a company can modify (eg the premium or benefit amount) or even abolish its pension scheme, it must reach agreement with its employees. For example, an employer must obtain approval from a majority of insured people, and from the trade union where unionised, when drafting the statute of an 'employees pension fund' (a traditional occupational scheme), as this requires equal contributions by employer and employee. In addition, if a company decides to replace its existing corporate pension plan with a DC scheme, it will need to obtain approval from the trade union, or from at least 50% of plan members where not unionised. The level of occupational pensions is basically determined within the system itself and is not directly influenced by other agendas that are dealt with in bargaining.

In the USA, 'traditional' DB pension plans have long been a key issue in collective bargaining - this is illustrated by the fact that in the private sector 69% of union members are covered by DB schemes, but only 14% of non-union workers. While many schemes are based on single-employer agreements, unusually for the USA a significant proportion are based on multi-employer collective agreements - multi-employer plans cover around 22% of those employees with a DB scheme, or 4% of the private sector workforce. The number and coverage of DB plans is declining, with many failing or being terminated. However, pensions remain a central issue in bargaining, and as a general rule companies have in recent years sought to reduce employer contributions while unions have sought to preserve or increase them (in non-union workplaces, employers can reduce contributions or change plans more or less at will, but not in a unionised environment). Collective bargaining appears to play little part in the DC schemes that are now becoming widespread.

With regard to the governance of occupational pension schemes, in the EU trade unions and employer organisations play a particularly important role in the Netherlands, Denmark and Sweden, where the systems are essentially based on sectoral collective agreements. In Denmark, for example occupational schemes are run by the social partners from the relevant sectoral bargaining unit, and employees sit on the funds' boards. Joint social partner management is also generally the rule in those countries where similar systems based on sectoral collective agreements are currently developing or being promoted - notably Belgium, Germany and Italy. Joint social partner governance also plays a key role in France's two national intersectoral supplementary pension schemes. Where occupational schemes are largely company-based, the involvement of the social partners in their governance tends to be much weaker - eg in the UK and Spain. This also appears to hold true in Japan and the USA. However, multi-employer DB plans in the USA are generally operated under the joint trusteeship of labour and management.

Occupational pensions - disputes

With pensions having become such a prominent and sometimes contentious topic, disputes and industrial action relating to pensions have arisen in a number of EU countries in the past few years. The most high-profile conflicts have been over government reforms of state pensions, with trade unions organising 'political' strikes in protest, along with major demonstrations and other expressions of opposition, in countries such as Austria (AT0306201N), France (FR0306104F), Greece (GR0312101N) and Italy (IT0311102N). While this kind of dispute between governments and unions, mainly over state pension provision, has become relatively widespread, industrial conflict over occupational pensions between trade unions and employers, especially in the form of strikes, has so far been less common. Where such action does take place, it is usually in those countries where occupational pensions form part of the general bargaining agenda. In such cases, it is difficult to separate pensions from other issues as the cause of a dispute. In the Netherlands, for example, there is a straightforward link between pay and pensions, with rising contributions for the latter increasingly cutting into space for the former. During a major private sector strike in Denmark in 1998 (DK9805168F), occupational pension contributions were one of a number of issues in dispute, alongside pay and leave, which were eventually traded off in a deal. Strikes specifically over occupational pensions issues and unrelated to the regular collective bargaining process seem very rare and cases are reported only from Ireland and the UK. Notably, pensions, and especially fighting final-salary scheme closures, have become a high-profile issue for UK trade unions (UK0301109F). In 2002, following the breakdown of talks between the ISTC trade union and the engineering firm Caparo Group, the UK's first strike over the closure of a final-salary pension scheme took place, and several similar stoppages have since been held or threatened at other organisations.

There have been a number of industrial disputes over occupational pensions in the USA in recent years (though disputes over healthcare coverage have been far more visible - see above), sometimes linked to other issues as part of overall collective bargaining. The issue may also become a source of conflict outside the normal bargaining process. For example, at the end of 2004 flight attendants at US Airways authorised their union to engage in strike-related activities should a federal bankruptcy court permit management to cancel the collective agreements of 20,000 employees and terminate the pensions of 53,000 existing and former workers (as well as cutting the health benefits of almost 11,000 retirees).

No industrial action over occupational pensions issues is reported from Japan. With regard to the state pensions system, in opposition to a 2004 pension reform, the National Confederation of Trade Unions (Zenroren) - the country's second-largest union centre - held a one-day strike, in which about 10,000 members participated. The last large-scale strike over pensions was in 1973, when several national centres jointly gathered 3.5 million members to demand the introduction of indexation.

Occupational pensions - social partner views

In the EU, the social partners take a wide variety of viewpoints on occupational pensions, reflecting the differing importance and development of such schemes. For example, in countries where occupational pension schemes have only recently started to develop, trade unions may fear that their introduction (or generalisation) will weaken the first pillar of state pensions, while some employers may fear that they will have to contribute to the schemes, either at all or too much. More generally, employers and unions may have different opinions on who should contribute how much to occupational schemes and for how long. They may disagree on the type of scheme, notably DC versus DB arrangements. Last but not least, they may disagree on whether occupational pensions should be seen as an individual employee benefit or should be based on collective agreements. To take the example of Belgium, the two main trade union confederations have in the past been united in their condemnation of occupational pension schemes, considering them a tool to weaken the first pillar. During consultations leading to a recent new law on complementary pensions, and after it came into force in 2004, the confederations have developed a more pragmatic stance, focusing on issues such as calling for a stronger social partner presence in the management of occupational funds. The main employer confederation supported the law, but is pressing for a greater individualisation of pension schemes. It believes that the occupational pension system created by the new law is close in some respects to a first-pillar system. The employers favour DC schemes, to keep pension costs visible and under control, and disagree with the restrictive conditions placed on individual pension schemes, which it believes should be a major human resource management tool.

Another example of the range of social partner positions on occupational pensions is Spain, where trade unions consider the (still limited) occupational pension system as a good supplement to the state pension in those cases where companies can afford it. Spanish employer organisations tend to favour private schemes within a mixed system, in which the public pension, without costing too much, will keep its central role. From the employers' viewpoint, occupational pension schemes, based on contributions, can fulfil a complementary role, particularly in binding key workers to the company. A topical issue in the UK is a trade union call for compulsory employer contributions to occupational pensions. The Trades Union Congress (TUC) wants to: force all companies to provide more substantial pension schemes, with a 10% contribution from employers and a 5% contribution from employees; make it compulsory for workers to join schemes; and for 'pension pay' to be given the same level of protection as current pay. Employers' bodies have lobbied strongly against compulsory employer contributions. An example of greater harmony is Denmark, where the social partners administer the sector-based system together, and see it as a joint project. In their reaction to a 2003 official report that called for more freedom in the choice of pension funds, the partners were unanimous that pensions should not be 'gambled' with, and that proposed greater freedom for workers to transfer pensions would undermine the element of solidarity in schemes.

One of the main points of pension contention in the USA is the DB or DC status of occupational schemes (see table 4 above). In DB plans, companies are obliged to continue making contributions regardless of the revenues earned on the plan's investment - even if revenues fall below levels needed to keep the plan adequately funded. With a DC plan, a company is responsible only for establishing the plan and deciding whether to match a percentage of employee contributions. The rise of DC schemes is thus not difficult to understand. Companies are free to select the type of pension plan they offer to employees, and the overwhelming majority of new companies are setting up DC plans, and older companies are switching from DB to DC. For their part, trade unions stress a need to preserve DB schemes and highlight the risks involved with participation in DC plans. AFL-CIO has been promoting in Congress a proposed item of legislation (the Employee Free Choice Act) that seeks to overcome the most serious obstacles to union organising and first-contract negotiations. AFL-CIO stresses the relevance to pensions - making it easier to organise a union will, it says, strengthen the collective bargaining system, which might over time bring more contributors to occupational pension plans. The Congress-directed activity of companies has included calling for changes that would let them make smaller pension contributions and smaller pay-outs when people retire. However, because employers are not legally obliged to provide pension coverage to their employees, the incentive to exert concerted political pressure in order to shape policy outcomes does not appear strong. Employers simply 'vote with their feet', and many are either not offering pension coverage or offering severely reduced coverage. In certain sectors, a good pension scheme is one way to attract and retain employees, but this positive aspect needs to be weighed against the financial cost of contributing to a plan incurred by the company.

In Japan, there seems arguably to be a fair degree of social partner consensus over occupational pensions issues. The newly established DC pension plan is aimed at enhancing pension portability for 'job switchers' and providing better conditions for employees. However, this pension scheme also, from a trade union viewpoint, contains some destabilising factors, in that it will transfer pension management risks from employers (or pension funds) to employees. When parliament passed the Defined-Contribution Pension Law in 2001, Rengo expressed its opposition, because 'the surrounding conditions are not yet appropriately prepared for such a pension programme to be introduced'. Employers' bodies basically supported the legislation.


It seems that 2003-4 was not a period of radical change in industrial relations in the EU, Japan and the USA, with the main feature being a continuation of previously established trends. However, there were signs of some developments that may point the way to longer-term change.

In Japan and the USA, collective bargaining coverage continued its long and relatively slow decline (mirroring falling trade union membership), now standing at around a fifth and an eighth of the workforce respectively. In the EU Member States, there was little change in bargaining coverage reported in 2003-4 (the longer-term picture is general stability in the EU15, though decline in a few cases such as the UK). However, at overall EU level the accession of the 10 new Member States in 2004 - most of which have much lower bargaining coverage rates than the old Member States - has reduced the coverage of the EU workforce by bargaining. Nevertheless, somewhere around two-thirds of the workforce of the expanded EU are probably still covered.

Bargaining remained almost entirely decentralised in Japan and the USA, and indeed Japanese unions experienced greater difficulties than in the past in even setting industry-wide wage increase targets in the 2003-4 enterprise-level bargaining rounds. In the EU countries, bargaining was, as usual, conducted at all levels from the national intersectoral to the individual workplace, depending on national systems, while still being considerably more centralised overall than in Japan and the USA. However, the new EU Member States generally have decentralised bargaining systems and enlargement has thus arguably shifted the 'average' level of bargaining towards the company level. During 2003-4 there was also increasing pressure towards decentralisation in a number of 'old' EU Member States - prime examples being France and Germany - with long-standing systems of sectoral collective bargaining (though generally the move is towards 'organised' decentralisation while maintaining a sectoral framework).

Despite differing economic environments, the pay outcomes of bargaining were generally moderate across the EU, Japan and the USA. Japanese unions continued to face difficult bargaining rounds, faced with a choice between maintenance of employment levels and pay rises in the context of a largely stagnant economy, and the resulting wage increases continued to be low. A new stress on pay increases for workers in SMEs produced some modest results. In the USA, economic growth and major productivity gains were not reflected in negotiated wage increases, which were lower than they had been during the 2001-2 recession. At overall EU level, average collectively agreed pay increases fell from 2002 to 2003 (though with substantial differences between countries), and the trend seems likely to have continued in 2004. In a number of countries pay moderation has been an explicit aim of central agreements. While the overall trend seems similar in the EU15 and new Member States, the average level of pay increases remains higher in the latter group (though the gap seems to be narrowing).

Working time was not, it seems, an important issue in collective bargaining in the USA. In Japan, the Rengo trade union confederation emphasised reduced working hours and work-sharing schemes, as well as eradicating unpaid overtime work, but there is little evidence as to whether it achieved its aims. In the EU15, with major working time reductions apparently largely a thing of the past, flexibility, in various forms, was the main theme in bargaining on working time (sometimes in exchange for small-scale reductions). In the new Member States, the active role of bargaining in setting normal weekly hours is relatively slight, especially in many central and eastern European countries, and collective agreements often do not deviate from the statutory normal hours. A novel feature of 2003-4 was that a number of company agreements in Germany and elsewhere provided for increased working time, often without any additional pay, in exchange for commitments on maintaining jobs. These cases generated much debate across Europe and called for an increase in agreed working time by a range of firms and employer organisations, which met strong trade union opposition.

The single most important issue in US collective bargaining remained health insurance, against a background of rising healthcare costs. In Japan, there were some signs of diversification beyond pay in the regular annual spring bargaining rounds. The bargaining agenda in some EU Member States (especially the 'old' ones) continued to encompass a wide and varying range of issues beyond pay and working hours.

With regard to employment and labour legislation, an interesting feature of 2003-4 was that there was considerable activity in Japan and the USA, where legislation generally plays a lesser role that in most EU countries. Notably, Japan seems to be increasing the volume and scope of its employment legislation to deal with new economic and social conditions, with a raft of adopted and proposed new measures. In the USA, one of the main items of basic employment legislation, the 1938 Fair Labor Standards Act, was amended, with more changes in the pipeline. The EU Member States saw their usual steady stream of new legislation covering all aspects of the employment relationship, though 2003-4 was a relatively quiet period in terms of the adoption and proposal of new EU employment and labour legislation.

The already comparatively low level of trade union membership and density continued to fall in the USA and Japan. In the EU (where the overall union density is considerably higher than in Japan and the USA), 2003-4 is likely to have witnessed a continuation of the trends of the past decade - rising membership (though not necessarily density) in a substantial number of countries, but an overall fall in union membership across the EU, mainly because of massive membership losses in some central and eastern European countries and significant losses in large countries with high absolute numbers of union members, such as Germany and the UK. Against this backdrop, union mergers continued in Japan and many EU countries, while proposals for a radical merger and restructuring process threatened to split the USA's AFL-CIO confederation. Another common theme has been for unions to reach out to new groups outside their traditional constituencies. On the employer side, a number of organisations restructured or merged in some EU Member States in 2003-4.

In Japan and the USA, labour disputes and industrial action remained at very low levels in historical terms. The picture was more mixed in the EU15, but the data available again indicate that a general trend towards lower levels of strikes and similar activity has been maintained in most countries in recent years. In some of the generally low-strike new Member States, however, 2003-4 saw a number of signs of increasing militancy, often over pay.

Turning to the special themes examined in this report, overall it appears that the extent to which migration has become an 'issue' in industrial relations depends very much on the level of labour immigration and the nature of national industrial relations systems. In relatively low-immigration Japan, the issue is subject to some consultation with the social partners at national level, but does not appear to feature at all in the decentralised enterprise-based collective bargaining system. Similarly, enterprise trade unions seem generally to have little interest in organising foreign workers, with the relatively little union activity in this area carried out by industry-level and community unions, which arguably have a wider perspective. By contrast, in the high-immigration USA, many unions are now giving priority to the interests of immigrant workers, seeing an alliance with them as having a potentially crucial role in reviving the trade union movement. This has led to a number of major initiatives and organising drives. However, here is little evidence of the extent to which issues of specific relevance to immigrant workers are taken up in (company-level) bargaining, and no explicit involvement of unions and business/employer organisations in policy-making migration issues is reported (unsurprisingly, given the general absence of social partnership or tripartite concertation). On the other hand, reflecting national social partnership systems, in most EU countries (which have varying levels of immigration) trade unions and employer associations are consulted by governments on issues relating to migrant workers. Trade unions in a minority of countries have launched initiatives aimed specifically at recruiting migrants, and support, assistance and campaigning seems relatively widespread. While the issue does not generally seem to be high on the collective bargaining agenda, there is some bargaining activity on migrant worker-related themes in many Member States, covering general matters such as non-discrimination and equal treatment, or more specific integration and support measures. This activity occurs largely in those countries with an important national/intersectoral and - especially - sectoral level of bargaining, while company-level bargaining on migration issues seems to be rare. This might suggest that more centralised bargaining systems are better equipped to deal with such matters.

The overall picture seems rather similar in the area of pensions, against a common background of concerns about the effects of demographic change. In most EU Member States, the social partners have a consultative or advisory (and in some cases management) role with regard to state pensions, while there is some social partner consultation in Japan and none in the USA. As for occupational pensions, there is an appreciable degree of collective bargaining on the topic in most 'old' EU countries. The extent of bargaining over occupational pensions in each of these countries depends on the relative importance of such schemes, and the involvement of social partners in them, but there appears to be a general tendency for bargaining - especially at sector level - to play an increasingly important role in this area, often encouraged to do so by recent legislation promoting occupational pension provision. Company-level bargaining also plays a role in some EU countries. In the USA, collective bargaining plays a key part in traditional defined-benefit occupational pension schemes, with the decline in the former reflected in the decline in the latter (which are often being replaced by defined-contribution schemes - a trend also seen in Japan and some EU Member States). Interestingly, this is one of the very few areas where multi-employer bargaining plays a significant role in the USA. In Japan, company-level bargaining (the only level) plays a role only where a scheme is to be established or the employer wants to make changes to an existing scheme.

Mark Carley, SPIRE Associates

Useful? Interesting? Tell us what you think. Hide comments

Eurofound welcomes feedback and updates on this regulation

Add new comment