Industrial relations in the EU, Japan, US and other global economies, 2005–2006

Printer-friendly version
  • Observatory: EurWORK
  • Topic:
  • Date of Publication: 14 November 2007

Roberto Pedersini

  [ pdf version size 519 kb ]

This report gives an overview of the main industrial relations developments in the European Union, Japan and the US in 2005 and 2006. The findings are the result of an initiative by the European Industrial Relations Observatory (EIRO), in the context of a research project started in 2000 to compare and ‘benchmark’ elements of industrial relations in these three major economic areas. The second part of the report investigates two elements of the social and economic environment which have recently gained greater relevance in the debate on the characteristics and impact of growing global competition: Temporary agency work (TAW) and small and medium-sized enterprises (SMEs). While the study mainly covers the EU Member States, Japan and the US, it also includes data on emerging economies, notably Brazil, China and India.

The study was compiled on the basis of individual national reports submitted by the EIRO correspondents. The text of each of these national reports is available below. The reports have not been edited or approved by the European Foundation for the Improvement of Living and Working Conditions. The national reports were drawn up in response to a questionnaire and should be read in conjunction with it.



Comparing three key economic areas such as the European Union, Japan and the United States is an interesting task. However, it is also somewhat problematic, particularly in terms of analysing institutional systems like industrial relations, which are deeply rooted in national traditions, rules and even cultures. The main difficulty is that in the case of many aspects of industrial relations – for instance, basic facts like trade union density and collective bargaining structure and coverage – the differences within the EU may be as large as, or even greater than, the differences between the EU, Japan and the US.

However, there are at least two reasons why such a comparison can be deemed useful from an industrial relations perspective. First, it allows for a comparison between industrial relations trends occurring in the EU and those taking place in Japan and the US. Such an analysis would be particularly relevant, for instance, in addressing questions on the possible ongoing ‘Europeanisation’ or even ‘Americanisation’ of industrial relations. Second, despite differences across Member States, the EU assigns a crucial role to industrial relations as a means for fostering social cohesion, while playing a unifying role, especially in the context of a continuously changing market.

Against this background, the following sections will review some basic industrial relations developments in the EU, Japan and the US in 2005 and 2006. While the comparison is by no means exhaustive, it nevertheless provides an interesting insight into industrial relations in these three leading economic areas.

Economic and employment context

In 2005 and 2006, the economic context in the EU, Japan and the US was generally favourable, despite the relative decrease in real gross domestic product (GDP) growth compared with that experienced in both the EU and the US in 2004. The similarities observed in 2005 and 2006 were, however, related to quite different trends: a slowdown in the US, economic recovery in Japan and a steady but limited growth in the EU (Figure 1). In Japan in particular, the recovery of the national economy has continued, with GDP growth reaching between 2.6% and 2.8% and with 2005 marking the end of a deflation cycle that had begun in 1999.

Figure 1: Real GDP growth in the EU, Japan and the US, 2000–2006

Figure 1: Real GDP growth in the EU, Japan and the US, 2000–2006

Source: European Commission, 2006b

As the findings presented in Table 1 show, employment indicators for these three areas were also positive, with signs of a slight increase in the proportion of employed persons and in employment rates and a small decrease in unemployment rates. In 2006, the employment rate in the 25 EU Member States (EU25) was 63.9%, compared with 69.3% in Japan and 71.5% in the US. In the same year, the unemployment rate stood at 8.7% in the EU, 4.4% in Japan and 5.1% in the US. Despite the general improvement in employment indicators from 2005 to 2006, the presence of significant labour market performance gaps between the EU, Japan and the US was confirmed.

A slight increase in the nominal compensation of employees was found in all three economies. Nonetheless, the real gains in compensation were below productivity growth, while a net decrease in real labour costs was observed, thereby contributing to the competitiveness of companies. These trends were more significant in Japan and only marginal in the US, while the EU took a more intermediate position in this respect.

Table 1: Macro economic indicators for EU, Japan and US, 2005–2006
  2005% 2006%
Real GDP (% growth)    
EU25 1.6 2.3
EU15 1.5 2.2
Japan 2.6 2.8
US 3.5 3.2
Inflation (% growth, CPI, harmonised for EU, national for Japan and US)    
EU25 2.2 2.1
EU15 2.1 2.1
Japan -0.3 0.7
US 3.4 2.9
Employment (%)    
EU25 63.3 63.9
EU15 64.7 65.3
Japan 68.7 69.3
US 71.2 71.5
Occupied population (% growth)    
EU25 0.9 0.9
EU15 0.7 0.8
Japan 0.4 0.4
US 1.8 1.4
Unemployment (%)    
EU25 9.0 8.7
EU15 8.0 7.9
Japan 4.7 4.4
US 5.5 5.1
Nominal compensation per employee (% growth)    
EU25 2.5 2.8
EU15 2.6 2.8
Japan 0.6 0.8
US 4.6 3.8
Real compensation per employee (% growth, private consumption deflator)    
EU25 0.3 0.7
EU15 0.5 0.7
Japan 1.3 0.3
US 1.7 1.5
Labour productivity (% growth)    
EU25 0.9 1.4
EU15 1.0 1.4
Japan 2.3 2.4
US 1.7 1.8
Nominal unit labour costs (% growth)    
EU25 1.5 1.3
EU15 1.6 1.3
Japan -1.6 -1.5
US 2.8 2.0
Real unit labour costs (% growth)    
EU25 -0.4 -0.5
EU15 -0.3 -0.5
Japan -0.3 -1.2
US 0.0 -0.5

Source: European Commission, 2006b

Legislative developments

A considerable number of legislative interventions in relation to labour issues occurred in this period, particularly in Japan and the EU. In the EU, some of these interventions were linked to the implementation of European directives and policies, although a significant proportion were introduced under the initiative of national actors, with several aspects of these changes having relevance from an industrial relations perspective. In the US, one notable development in the legislative field relates to the pensions issue.

In Japan, a number of significant legislative developments took place in 2005 and 2006. The revised Equal Opportunity Law on securing equal opportunity and treatment between men and women in employment was passed in June 2006, coming into effect in April 2007. This legislation includes provisions prohibiting gender discrimination, including indirect discrimination, and unfair treatment in the event of maternity leave. The law will be implemented through specific measures to be defined through an Ordinance of the Ministry of Health, Labour and Welfare. In the summer of 2006, a discussion commenced on revising the Part-Time Work Law – focusing in particular on the issues of wages, access to education and training, and welfare benefits – with a view to reviewing the differential treatment of part-time workers and regular workers. As was the case in the previous revision in 2003, it is likely that the debate will focus on whether the law should ensure ‘equal’ or ‘balanced’ treatment for part-time workers. Whereas the 2003 revision opted for a ‘balanced treatment’ approach, the present review seems to be more oriented towards the implementation of an ‘equal treatment’ principle.

In April 2006, the Law concerning temporary measures for the promotion of shorter working hours was abolished and the Law on special measures to improve the setting of working hours entered in force. In accordance with the new rules, a labour management committee should be established to define working hours and to promote new systems aimed at organising working time. Concerns have been raised that these new provisions may lead to longer working hours. In response to such concerns, a parallel revision of the Occupational Health and Safety Law now requires that employees who work more than 100 hours a month – above the 40-hour weekly threshold – should report to an occupational health physician. In addition, talks on the revision of the Employment Measure Law have started. The revised text, due to be discussed in parliament in 2007, should contain measures to foster the employment of young people, women and older workers, in light of the growing labour shortages; it should also provide for the establishment of a reporting system for foreign workers and trainees, aimed at preventing company violations in the treatment of such workers.

In the US, a major legislative development from an industrial relations perspective was the enactment of the Pension Protection Act in the summer of 2006. This legislation enables employers to increase the amount it deducts from employees’ wages to finance pension plans. It is expected that the new piece of legislation will favour a shift from the traditional employer-funded or ‘defined benefit’ pension plans to ‘defined contribution’ schemes, which invest on the stock market. Since pension plans are a crucial element in collective bargaining, the pressure which may derive from this law at the bargaining table could weaken pay increase demands by trade unions, instead encouraging deals that aim to preserve pensions.

Another important development concerns a decision taken by the National Labor Relations Board (NLRB) in late September 2006 (Oakwood Healthcare, Inc, Case 7–RC–22141 (298Kb PDF), usually referred to as the Kentucky River case). The decision, based on a controversial 3–2 majority, established that the performance of ‘supervisory tasks’ for at least 10% to 15% of individual working time would result in an employee acquiring a supervisory status under the meaning of section 2(11) of the National Labour Relations Act (NLRA) (650Kb PDF). This would entail the exclusion of the concerned worker from NLRA protection, including trade union membership and the associated bargaining coverage. Such a decision may have a substantial impact on higher-skilled occupations and in sectors such as healthcare and construction, where limited supervisory tasks are often performed by a greater number of skilled workers. According to an estimate by the American Federation of Labor and Congress of Industrial Organizations (AFL-CIO), as many as eight million workers may fall into this category and be excluded, by law, from trade union representation and protection.

In the EU, the various Member States passed a number of relevant legislative measures in 2005 and 2006. Some of these were linked to the implementation of EU directives, as was the case in relation to the Directive establishing a general framework for equal treatment in employment and occupation (Directive 2000/78/EC (129Kb PDF)), which led to the introduction of important measures. The countries in which new regulations on equal treatment were passed include: Germany, which introduced a reform law covering equal treatment on all grounds; Greece, France and Spain, which enacted new rules on gender equality; and the UK and Estonia, which introduced provisions aimed at combating age discrimination. In addition, the Directive on informing and consulting employees (Directive 2002/14/EC) required some of the Member States to adopt new legislation in 2006 – namely, Bulgaria, Estonia, Ireland, Malta, Poland and Romania. Moreover, in order to comply with the EU acquis communautaire in the field of employment regulations before joining the EU on 1 January 2007, Bulgaria and Romania were required to enact a number of legislative texts. Apart from EU-led legislation, interventions in the various Member States have touched on virtually all dimensions of employment and labour market regulations, from atypical work, to social security systems and working time (TN0601103F, TN0703019S).

Finally, at EU level, major developments have included the enactment of the long-debated Services Directive (Directive 2006/123/EC (212Kb PDF)) and the new Equal Opportunities Directive (Directive 2006/54/EC (130Kb PDF)), which incorporated the existing Directives on the equal treatment of women and men, equal treatment in occupational social security schemes, equal pay and the burden of proof in cases of discrimination based on sex. Other initiatives concerned the possible revision of the Working Time Directive (Directive 2003/88/EC) and the portability of supplementary pension rights.

Social partner developments

Trade union mergers and splits

In the EU in 2005 and 2006, a number of significant talks and trade union mergers took place, particularly in the Nordic, central and eastern European countries, and in Austria and the UK. Among the many examples, some notable cases are worth mentioning:

  • in Sweden, the trade union IF Metall was established through the merger of the Swedish Metalworkers’ Union (Svenska Metallarbetareförbundet, Metall) and the Industrial Labour Union (Industrifacket, IF);
  • in Denmark, the United Federation of Danish Workers (Fagligt Fælles Forbund, 3F) was founded on 1 January 2005, following the merger between the General Workers’ Union (Specialarbejderforbundet i Danmark, SiD) and the National Union of Female Workers (Kvindeligt Arbejderforbund, KAD);
  • in Romania, a new federation was established in 2005 – namely, the Federative Alliance of Civil Servants Trade Unions Sed Lex (Alianţa Federativă a Sindicatelor Funcţionarilor Publici Sed Lex, Alianţa Sed Lex) – merging two federations and becoming the second most important trade union confederation in the country;
  • in Slovenia, five Slovenian public sector unions came together in 2006 to establish the Confederation of Public Sector Trade Unions (Konfederacija sindikatov javnega sektorja, KSJS);
  • in Austria, three new trade union organisations were established through the mergers of seven affiliates of the Austrian Trade Union Federation (Österreichischer Gewerkschaftsbund, ÖGB) in 2006;
  • in the UK, the University and College Union (UCU) was created through the merger of two trade unions representing academic staff in the further and higher education sectors; meanwhile, Amicus and the Transport and General Workers’ Union (TGWU) continued the merger process, after Britain’s General Union (GMB) pulled out of talks, leading to the establishment of Britain’s largest trade union on 1 May 2007 – Unite.

These steps were taken with a view to increasing the coordination between trade union organisations – at peak or sectoral levels – and to facing decreasing membership and growing organisational costs (TN0601103F, TN0703019S).

In the US, a major development took place in 2005: namely, the split within AFL-CIO – the principal trade union confederation since the merger of the previously separate American Federation of Labor and the Congress of Industrial Organizations in 1955. Following disagreement on trade union policies, particularly in respect of organising non-unionised workplaces and sectors, a new federation called Change to Win was established. Formerly a coalition set up to support reform within AFL-CIO, the entity evolved into a new organisation after two of the largest member unions of AFL-CIO announced their disaffiliation on the eve of the 50th federation convention. The latter split involved the Service Employees’ International Union (SEIU), the then largest AFL-CIO trade union with 1.8 million members, and the International Brotherhood of Teamsters (IBT), the third largest union with 1.4 million members. The two unions left AFL-CIO in July 2005 and in late September of the same year formed the Change to Win alliance, together with five other founding unions, bringing together some six million members. The other members of the Change to Win alliance are: the United Food and Commercial Workers’ International Union (UFCW), which left AFL-CIO just after SEIU and IBT; Unite Here, which ended its affiliation in mid-September 2005; the Laborers’ International Union of North America (LIUNA) and the United Farm Workers of America (UFW), which both left in 2006; and the United Brotherhood of Carpenters and Joiners of America (UBC), the only trade union already outside of AFL-CIO, having left in 2001. Despite this split, AFL-CIO remains the largest trade union federation in the country, counting 55 affiliated organisations and some 10 million members.

In Japan, the decline in the number of trade unions continued – a trend which started at the turn of the 1980s: as at 30 June 2005, some 61,200 unions were recorded compared with almost 68,750 unions in 2000, which represents a decrease of about 11% (Japan Institute for Labour Policy and Training, 2007).

Trade union membership decline

In terms of trade union membership, the decline in union membership and density continued in both the US and Japan. In the US, according to the Current Population Survey, almost 15.4 million trade union members were recorded in 2006, representing 12% of all wage and salary workers (Figure 2). Since 2000, trade union density decreased by nearly 1.4 percentage points and membership by almost one million workers.

Figure 2: Trade union membership (thousands) and density (%) in the US, 2000–2006

Figure 2: Trade union membership (thousands) and density (%) in the US, 2000–2006

Source: US Department of Labor, Bureau of Labor Statistics, 2000–2006

In Japan, the number of trade union members was slightly above the 10 million mark in 2006, corresponding to an estimated trade union density level of 18.2% (Figure 3). Compared with figures for 2000, trade union membership had dropped by 1.5 million workers and union density levels by about 3.3 percentage points. In fact, all of the main trade union confederations in Japan lost members: according to data for 2005, when the estimated unionisation rate stood at 18.7%, the Japanese Trade Union Confederation (Rengo) had 6,543,000 members (down by 51,000 members compared with 2004), the National Confederation of Trade Unions (Zenroren) counted 723,000 members (22,000 members less than in 2004), while the National Trade Union Council (Zenrokyo) had 150,000 members (a decrease of 4,000 members compared with the previous year).

Figure 3: Trade union membership (thousands) and density (%) in Japan, 2000–2006

Figure 3: Trade union membership (thousands) and density (%) in Japan, 2000–2006

Source: Ministry of Health, Labour and Welfare, Basic Survey on Labour Unions (data available on the Japan Institute for Labour Policy and Training website), 2000–2006

A similar downward trend in trade union membership, albeit at generally higher levels and less pronounced, can be traced in the EU. However, the picture is far more varied across the EU Member States, with trade union density ranging from between 70% and 80% in some Nordic countries to less than 10% in France (European Commission, 2006a) (TN0403105U, TN0207104F). The main problems faced by trade unions appear to relate to recruitment difficulties in the private service sectors as well as in organising a more diversified workforce; the latter is attributed to the increasing presence of certain groups of workers, for instance women, atypical workers, young and older workers, migrants and those working in small and medium-sized workplaces (European Commission, 2006a). However, a number of efforts are being made to try to counterbalance this decline and strengthen membership. ‘Union revitalisation’ efforts include a greater emphasis on ‘grassroot unionism’ and partnerships with community or other non-governmental organisations (NGOs), a trend mostly observed in the US; such efforts also encompass new initiatives in terms of services and protection, especially those devoted to specific groups of workers, a trend more often found in the EU (Milkman and Voss, 2004; Frege and Kelly, 2004; Turner, 2007).

Employer organisations

In Japan and the US, no significant developments have taken place in the domain of employer organisations. The Japan Business Federation (Nippon Keidanren) remains the major Japanese employer organisation, with 1,346 member companies, 130 major national sectoral organisations and 47 regional organisations (May 2006). In the US, there is little evidence of any employer organisation playing an active role in industrial relations.

In the EU, some developments have taken place in this area, including mergers and the creation of new organisations. For instance, in Bulgaria, the two confederations – the Employers’ Association of Bulgaria (EABG) and the Bulgarian International Business Association (BIBA) – merged in 2006 to form the Confederation of Employers and Entrepreneurs in Bulgaria (CEEB). In Romania, a new employers’ organisation was established in the agriculture sector in 2005 – namely, the Major Agricultural Producers’ Organisation (Patronatul Marilor Producători Agricoli din România, PMPA) – in addition to the Employers’ Organisation of Young Entrepreneurs (Patronatului Tinerilor Întreprinzatori din România, PTIR). In Norway, a new employer association was founded at sectoral level in the manufacturing sector in 2005 – called the Federation of Norwegian Industries (Norsk Industri) – as a result of a merger between the Federation of Norwegian Manufacturing Industries (Teknologibedriftenes Landsforbund, TBL) and the Federation of Norwegian Process Industries (Prosessindustriens Landsforening, PIL). In Denmark, in 2006, a merger took place between the Danish Commerce and Service (Dansk Handel og Service, DHS) employer organisation and the Danish Chamber of Commerce (Handel, Transport, Service, HTS-I) trade association, resulting in the establishment of the Danish Chamber of Commerce (Dansk Erhverv, DE). In Portugal, attempts to create single peak employer organisation began in 2005, but were suspended in 2006. There was greater cooperation between the different associations in Lithuania and the Czech Republic in 2006.

Collective bargaining

It is difficult to identify common trends in collective bargaining across the three economies. In general, it could be said that, in recent years, collective relations have weakened in Japan and particularly in the US, as reflected by the significant reduction in unionisation illustrated above. In the US, survey data suggest that difficulties in concluding collective agreements had somehow increased in the early 2000s compared with the mid-to-late 1990s, with some 10% of negotiations ending without agreement (Cutcher-Gershenfeld et al, 2007). In fact, in both countries, due to a collective bargaining structure which is essentially based on company or plant-level agreements, coverage rates are close to trade union density levels. In these countries, the fundamental difference in terms of the scope of industrial relations is actually between unionised and non-unionised workplaces. In the US, in 2006, the collective bargaining coverage rate was slightly higher than unionisation and amounted to 13.1% or 16.8 million workers, including non-unionised members whose terms of employment are covered by collective agreements. In Japan, the coverage rate is likely to be around 20%, similar to the country’s trade union density level.

As far as the EU is concerned, no relevant changes took place in the existing structure of collective bargaining, which in many Member States is centred around industry-wide agreements, rather than bargaining at plant level as in Japan and the US. Some limited tendencies towards decentralisation were observed, for instance in the Netherlands and Greece in 2005; such a trend has been common in the relatively centralised bargaining system of continental Europe in recent years and has sometimes been associated with weakening industrial relations. However, these tendencies towards decentralisation have also been accompanied by a strengthening of collective bargaining in some central and eastern European countries – both at company level (in Estonia and Romania) and even at sectoral level (in Slovakia). This is probably a reflection of how collective bargaining remains a key regulatory tool of labour relations in Europe, despite some signs of erosion of bargaining coverage in Germany and the UK (TN0703019S).

In Europe, the traditional distinctions remained intact: that is, between the Nordic and continental European countries, where multi-employer bargaining prevails, and the Anglo-Saxon and central and eastern European countries, where decentralised company-level bargaining is the dominant pattern. In some instances, when combined with various extension mechanisms, multi-employer bargaining ensures much higher collective bargaining coverage rates of almost 100% – as seen in Austria and Belgium. On the other hand, the prevalence of decentralised bargaining entails significantly lower coverage rates, as is the case in the UK and in many central and eastern European countries. However, it may be worth noting that, even in such instances, collective bargaining in the EU usually affects about one third or more of the total workforce – a level which is substantially higher than those found in the US and Japan. Only in a couple of the Baltic countries, namely Latvia and Lithuania, were coverage rates of between 10% and 20% recorded in the early 2000s (TN0207104F).

Main themes of collective bargaining

In terms of topics covered by collective bargaining, wage increases and economic aspects continued to be at the centre of the negotiations. However, health coverage also remained a crucial issue in the US, due to the rising costs of health insurance and the attempts by many employers to reduce their health plan costs. As for trends, the abovementioned survey data (Cutcher-Gershenfeld et al, 2007) show, between 1996 and 2003, a significant reduction in the percentage of collective agreements consisting of wage and benefit increases, falling from well above 90% to slightly over 80% in terms of wage increases and from around 70% to less than 50% in relation to benefit increases. At the same time, innovative aspects – such as measures concerning job security, worker input, team-based work systems and joint committees – seemed to decline, involving at most only around 20% of all agreements; only new pay systems maintained an upwards trend in collective bargaining agreements. While the authors cannot rule out the influence of a short-term economic downturn on the 2003 findings, the overall picture underlines the difficulties in relation to collective bargaining faced in the US in the early 2000s.

In Japan, for which substantial data are available for trade unions affiliated to Rengo, negotiations in individual companies covered a wide variety of topics, such as: working time reductions, which were requested by 2,267 unions, of which 701 eventually reached agreement; the creation of company minimum wages (2,750 demands and 1,221 agreements); the possibility to remain at work after the age 60 years (2,725 demands and 1,106 agreements); job security (1,695 demands and 824 agreements); and work–life balance (1,462 demands and 662 agreements). Moreover, efforts to establish coordinated demands in the SME sector continued through the Joint Struggle for Small and Medium-sized Enterprise Unions. A similar initiative was also launched through the establishment of the Part-time Worker Joint Struggle Council, which aims to coordinate the demands of sectoral trade unions participating in company negotiations, to ensure wage improvements and effective equal treatment for part-time workers.


As already mentioned, wage increases and economic aspects continued to remain at the centre of collective bargaining in the three economies under consideration. In Japan, the ‘wage offensive’ (shunto) in the spring of 2006, owing to the country’s improved economic situation, represented a new departure from the somewhat ‘passive’ demands for homogeneous basic wage increases across all sectors, which had been put forward in the previous five years. A new ‘wage improvement’ strategy was launched which, given the relatively good performance of industries and companies, leaves significant scope for and encourages different demands, for instance, in terms of higher wages for young workers or higher allowances for overtime work. The results in terms of springtime wage increases were pay rises of 1.71% and 1.79% in major enterprises and of 1.37% and 1.47% in SMEs in 2005 and 2006 respectively. For both types of enterprises, these upward trends consolidated the reversal which took place in 2004, after more than a decade of diminishing pay rises, linked to the combined impact of economic difficulties and deflation. Nonetheless, with the end of deflation, the most recent real pay rises remain below the levels reached in the early 2000s.

In the US, wages in unionised workplaces remain significantly higher than those in non-unionised workplaces, owing to the lack of collective bargaining coverage in the latter. In 2006, the average weekly pay of trade union members was almost 30% higher than that of non-union workers, at USD 833 compared with USD 642 respectively (Figure 4). In addition, annual increases in the average weekly pay have been consistently higher for union members since 2002, thereby leading to a widening pay gap.

Figure 4: Average weekly wages of union and non-union workers in US (USD) and % increase, 2000–2006

Figure 4: Average weekly wages of union and non-union workers in US (USD) and % increase, 2000–2006

Source: US Department of Labor, Bureau of Labor Statistics, 2000–2006

In the EU, the results in terms of collectively agreed pay rises were quite diverse across the Members States in 2005 and 2006. The nominal average collectively-agreed pay increase in the EU25 reached 4.4% in 2005 and 5.2% in 2006. However, it should be added that these summary indicators cover a range of different pay increases, varying from under 2% for both years in Germany and the Netherlands to well above 10% in certain Baltic and eastern European countries. In real terms, the EU average wage increases amounted to 1.7% and 2.6% respectively for 2005 and 2006. However, the variation between countries does not substantially diminish when considering real developments: wage moderation was especially evident in the EU15 countries, with real wages increasing on average by 0.6% and 0.8% in 2005 and 2006 respectively; significantly higher real pay rises were observed in the new Member States that joined the EU in 2004, amounting to 3.5% and 5.3% in the two reference years, but with the significant exception of Hungary and Slovenia (TN0704029S).

Industrial conflict

In relation to industrial conflict, both the US and Japan showed limited and declining levels of conflict. In the US, despite strike initiatives which drew the attention of the wider public, such as the two-day New York transport strike in December 2005, conflict remained relatively low, with some 100,000 workers involved in work stoppages in 2005 and 190,000 workers taking part in 2006. In Japan, the reduction in conflict levels continued, as reflected by all of the major indicators: the number of disputes dropped from 737 disputes in 2004 to 708 in 2005; disputes involving strikes fell from 173 disputes in 2004 to 129 in 2005; the number of days lost dropped from 9,755 to 5,629 days in 2004 and 2005 respectively; while the number of workers involved fell from 55,174 workers in 2004 to 27,295 workers in 2005. Such reductions represent a substantial decrease which, of course, may reflect contingent factors, but seemingly confirm the downward trend observed in the most recent years.

In the EU, industrial action followed the usual cycles of bargaining rounds in the different countries and was therefore linked to differences between employers and trade unions over the terms of agreement renewals – as was the case in Cyprus, Finland, Germany, Hungary and Italy. A certain proportion of the industrial action was also connected to public policy reforms – in the field of pensions in particular and the welfare state in general – as seen in Belgium, France, Greece, Italy and Slovenia. The most prominent strikes took place in parts of the public sector undergoing reform, such as: the civil service sector in Germany, Hungary, Norway, Portugal and the UK; the health sector in Cyprus, the Czech Republic, Germany, Hungary, Poland, Romania, Slovakia and Spain; the public transport sector in Belgium, Bulgaria, Estonia, France, Italy, Malta, Portugal and Spain; the education sector in France, Greece, Malta, Portugal, Romania and the UK; and the postal sector in Belgium, Malta and Poland. In the private sector, the most high-profile disputes at company level were often connected to restructuring activities, for instance, involving the following companies: BASF, Volkswagen, Fortis, Sonaca and Inbev in Belgium; SAS airlines in Denmark; AEG in Germany; Irish Ferries in Ireland; Nedcar in the Netherlands; K-Tel in Poland; and Opel in Portugal (TN0601103F, TN0703019S). In general, conflict levels have remained quite low and in many cases seem to show a declining trend.


Cutcher-Gershenfeld, J., Kochan, T., Ferguson, J.P. and Barrett, B., ‘Collective bargaining in the 21st Century: A negotiations institution at risk’, in Negotiation Journal, July 2007, pp. 249–265.

European Commission, Employment in Europe, Luxembourg, Office for Official Publications of the European Communities, 2006b.

European Commission, Industrial relations in Europe 2006 (1.44Mb PDF), Luxembourg, Office for Official Publications of the European Communities, 2006a.

Frege, C. and Kelly, J. (eds.), Varieties of unionism: Strategies for union revitalisation in a global economy, Oxford, Oxford University Press, 2004.

Japan Institute for Labour Policy and Training, Japanese working life profile 2006–2007, Tokyo, Japan Institute for Labour Policy and Training, 2007.

Milkman, R. and Voss, K. (eds.), Rebuilding labour: Organising and organisers in the new union movement, Ithaca, New York, Cornell University Press, 2004.

Turner, L., Beyond the workplace: The politics of labour and society, presentation at the International Industrial Relations Association (IIRA) 8th European Regional Congress, Manchester, 3–6 September 2007.

Adaptability in a global context: Temporary agency work and small and medium-sized enterprises


Recent economic and institutional transformations have exposed industrial and employment relations systems to ‘global’ challenges. Even if such systems remain substantially rooted in national settings, they are becoming increasingly open to cross-national comparisons and benchmarking. Policymakers, practitioners and economic and social actors no longer consider their specific national systems as fixed; rather, they often look for information on other systems, where the possible competitors or partners of domestic companies operate, as well as data on possible trends in transnational regulation. Employment and industrial relations constitutes an interesting field of research, which has attracted growing attention from researchers worldwide. Apart from theoretical and analytical issues, comparative analyses can contribute to addressing practical concerns, such as the identification of good practice and the definition of new policies.

For these reasons, it is interesting to compare the employment and industrial relations situation of the Member States of the European Union (EU) with those of the two other major ‘global players’ – Japan and the United States (US). Comparisons with the leading emerging economies, in particular those of Brazil, China and India, are also worthwhile, as these countries are likely to have a significant influence on the future transformation of the international economic environment. Two quite different types of comparisons emerge in this context: in the case of the US and Japan, the similarities, although not necessarily predominant, are certainly substantial, for example with respect to economic and social conditions. Conversely, in the case of the emerging economies, the differences outweigh the similarities. However, both comparisons can provide a greater insight into the global context in which European actors are likely to be operating in the near future, contributing to a mutual understanding and learning process.

Objectives of study

Since 2000, the European Foundation for the Improvement of Living and Working Conditions has, through the European Industrial Relations Observatory (EIRO), been carrying out a research project to compare and ‘benchmark’ elements of industrial relations in the EU, Japan and the US. Moreover, since 2004, the Foundation has strengthened this effort by extending coverage to other parts of Asia as well as Latin America, namely to Brazil, China and India.

This report aims to investigate two elements of the social and economic environment which have recently gained greater relevance in the debate on the characteristics and impact of growing global competition: temporary agency work (TAW) and small and medium-sized enterprises (SMEs). Both of these elements represent evolving features of the broad reorganisation process, which has mainly affected the advanced western economies since the economic shocks and recession of the 1970s. In particular, they have brought to the forefront the issue of the adaptability, or as it is more often referred to, the ‘flexibility’ of enterprises as a central focus of economic and political actors. More generally, TAW and SMEs have placed the spotlight on the ability of national economic systems to cope with more dynamic, competitive and internationalised markets.

In this context, TAW refers to the degree of labour market flexibility and, perhaps more specifically, to the possibility that companies can adjust their employment levels to short-term variations in demand or to unexpected circumstances, thereby supporting increased adaptability. At the same time, the growing relevance of SMEs may be considered a possible consequence of the restructuring and reorganisation processes linked to increased competition – a key component of globalisation. This has often led to the displacement and downsizing of large and vertically-integrated companies. Since the 1980s, large companies have resorted to outsourcing services, thereby contributing to the creation of a dense network of smaller companies integrated into the supply chain. Moreover, SMEs may be seen as a structural feature, which in itself can contribute to the flexibility of the economic system, due to the inherent adaptability of smaller organisations and their less formal and bureaucratic environments.

Naturally, the specific relevance and significance of TAW and SMEs can vary greatly in the different economies, depending on both institutional and traditional features. This may be especially true in the context of the particular focus of this study, namely labour and industrial relations. Thus, caution should be taken in making comparisons and special attention should be paid to the particular context in which TAW and SMEs are embedded. Focusing mainly on the advanced economies of the EU, the US and Japan, and partly on the emerging economies of Brazil, China, and India, the study will provide some clear illustrations of the diversity and complexity of national situations which derive from the interplay between other characteristics of the social, economic and institutional systems.

In relation to labour and industrial relations, TAW and SMEs present some specificities. TAW implies a triangular relationship between the temporary work agency, which is the employer, the employee and the user company. Therefore, employment and working conditions are based on a combination of different contractual conditions – involving the employment contract and supply contract – as well as a combination of organisational and material environments in which the work is performed. Similarly, workers’ rights, including representation and collective bargaining, involve, in principle, both the employer and the user firm.

Moreover, employment conditions in a broad sense encompass the structural element of being temporarily integrated in potentially different and continuously changing workplaces, with a possible emerging demand for some sort of continuity, including career prospects, even if across companies rather than within companies. As a result, TAW raises a number of questions relating to the legal and collective regulation of this triangular employment relationship, as well as to the presence of employment policies, but also of company policies, which aim to create the necessary conditions for the stable participation and career advancement of temporary agency workers in the labour market.

With respect to SMEs, labour relations are often characterised by a low level of formality and a certain degree of ‘paternalism’, involving dense personal relations that cut across a relatively lowly-differentiated hierarchy and the possible direct presence of the entrepreneur or general manager on the shop floor. Of course, these features are usually directly correlated with the actual size of the company. Such characteristics are mainly present in micro-enterprises and are progressively diluted as size of the company increases and as the organisation becomes more formalised. As for industrial relations, the trade union presence and collective bargaining are strongly related to the company’s size, although the strength of this link also depends on the sector in which the company operates, with collective representation usually being weaker the smaller the enterprise (see Foundation study on Employment relations in micro and small enterprises in the EU – Literature review (171 Kb PDF)). Therefore, the main questions in this case refer directly to the characteristics of labour relations, on the one hand, and to the role of industrial relations, on the other hand.

The report will address these two issues in separate sections, finishing with a general overview in the commentary section at the end. The main data sources for this study have been obtained from the EIRO overview on TAW and the EIRO comparative study on employment relations in SMEs. The report also refers to the responses obtained from a questionnaire sent out to EIRO’s international correspondents.

The EIRO report on TAW was commissioned by the Sectoral Social Dialogue Committee Temporary Agency Work and is based on an EIRO comparative study of TAW in the enlarged EU. The project involved a questionnaire survey administered through the EIRO network of national centres and affiliated institutions. Replies were received from 28 countries in total. These findings were further supplemented by additional data and detailed comments received from the social partners at both national and European levels (the European Confederation of Private Employment Agencies (Eurociett) and UNI-Europa), which were closely involved in the production of both the national and the overview reports from the very outset. The EIRO comparative study on employment relations in SMEs was compiled on the basis of individual national reports submitted by the observatory’s national centres. The study examined employment relations in SMEs in 16 EU Member States – namely, Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, the Netherlands, Poland, Slovakia, Slovenia, Spain, Sweden and the UK – in addition to Norway. Information on Brazil, China, India, Japan and the US was collected through a questionnaire which, following a set of questions on recent political, economic and industrial relations developments, included the main items prepared by the EIRO national centres for the purposes of the two comparative reports on TAW and employment relations in SMEs.

Temporary agency work

Factors driving demand and supply

A crucial issue in the analysis of TAW is how different institutional contexts promote a certain balance between the demand for adaptability by companies – in terms of what can largely be described as a form of numerical flexibility (see Goudswaard and de Nanteuil, 2000) – and the demand for fair employment conditions by workers, including the continuity of employment beyond the structural limitation of individual assignments. The latter could be achieved if TAW contributed to a substantial increase in personal employability, therefore improving subsequent employment and career prospects.

This dilemma, which has been analysed in the ‘debate on how best to balance employment generation and flexibility, on the one hand, with employment protection and security, on the other’ (Arrowsmith, 2006), can be addressed by taking into account the issue of discrimination – or, conversely, equal treatment – between temporary agency workers and permanent workers in comparable jobs within the user companies. Such a focus has been at the centre of the discussion in the EU on a possible harmonisation of TAW rules across Member States through the introduction of a European Commission directive.

From a more neutral point of view, it is possible to refer in general to ‘differential features’, with a view to identifying the possible comparative advantages for both the user company and the temporary agency worker of TAW compared with other contractual relationships or employment situations in the labour market. As for the diffusion of this contractual relationship, it can be assumed that the level of demand for and offer of TAW will depend on the degree and relevance of such comparative advantages. It is important to underline that such comparative advantages can only be partly connected to the ‘differential treatment’ of the workers involved and therefore to something that could possibly be labelled as ‘discrimination’. In fact, offer and demand can be reasonably linked to specific features of TAW, rather than to a ‘protection gap’ only. Thus, it is probably best to start the discussion by identifying such specificities.

A number of specific characteristics may favour the use of TAW ahead of other employment contractual relationships available to companies (Table 1). Firstly, TAW seems to respond particularly well to an immediate demand for labour on a short-term basis, as it allows companies to save on employee recruitment costs. The shorter the assignments, the more relevant these benefits are. Secondly, TAW may be used instead of ordinary selection processes, especially if the assessment of candidates can be best performed directly ‘on the job’, rather than through the analysis of CVs and through interviews. In practice, therefore, TAW may emerge as the preferred option for companies demanding temporary work if the implicit and explicit costs of search and selection processes are deemed excessive, due to the immediate need for labour, the short duration of assignments, or the types of positions to be filled. In such instances, the default alternative – namely, fixed-term work – may be deemed less favourable because it involves such costs. The influence of these components on the demand for TAW can also be linked to ‘structural’ variables, such as the composition of labour demand and offer in terms of job positions (which may in turn depend on other variables such as organisational models and types of technology) and sectors, or a combination of both of these factors.

A similar line of reasoning may apply to workers. TAW may be more preferable to workers if they have a short-term commitment to employment, possibly with a view to reconciling work with other interests or activities, and would like to have the option of renewing such a commitment at will; at the same time, they may wish to benefit from the earnings and accumulation of work experience provided by this form of employment (Table 1). In more specialised niches, TAW may also enable workers to gain a diverse range of experience in different organisational contexts, possibly with a view to entering into a professional occupation – for example, that of IT technicians. In the case of the ‘reconciliation preference’, workers’ preference for TAW can be linked to structural variables, such as age and gender, which can exert an influence on the long-term commitment to work and on the need for reconciliation. For instance, students and possibly pensioners and women may be likely candidates for TAW posts. In the case of the ‘experience preference’, the relevant ‘structural’ variables are likely to be similar to those outlined for companies: namely, the composition of labour offer in terms of job positions and sectors, or specific combinations of both. However, while a greater demand for TAW by companies may be linked to both high-skilled and low-skilled positions – with a significant presence of the latter – for workers, a higher offer of TAW is more likely to be linked to high-skilled occupations, for which diverse experience may be particularly relevant.

Other components affecting the demand for and offer of TAW may be related to the existing differences with other employment contracts and occupational situations. It could be expected that demand will escalate as the advantages of TAW increase for companies in terms of contractual and working conditions, compared with dependent employment (both open-ended and fixed-term) – or, conversely, as the ‘discrimination’ of temporary agency workers grows.

At the same time, workers will have a higher propensity to accept TAW if such employment contracts offer them advantages in terms of earnings and occupational prospects compared with unemployment benefits, if available, or the possibility of a more secure position if they were willing to prolong their job search. Thus, occupational policies and labour market conditions are crucial for encouraging the take-up of TAW positions, even if such choices are simply linked to the lack of other job opportunities. The latter scenario may involve prospective workers who find it particularly difficult to obtain a standard employment contract, or any kind of job for that matter, and who turn to TAW to escape unemployment. In this sense, TAW can provide work opportunities to disadvantaged segments of the labour market, including those who are long-term unemployed. Besides formal employment contracts, informal and irregular employment may, under certain circumstances, also arise as a viable alternative to TAW for both companies and workers, thus further complicating the overall picture. In general, it is possible to say that all of these ‘differential-based’ components regarding the demand for and offer of TAW are linked to ‘institutional’ variables and notably to the legislative regulation of employment contracts, and possibly also to collective bargaining, as well as to the characteristics of employment policies and of labour markets.

Table 1: Factors influencing demand for and offer of TAW
  Demand for TAW by companies Offer of TAW by workers
Specific advantages (structural variables: job position, sector, age, gender) Short-term demand Savings on recruitment costs Short-term, unstable commitment to employment Reconciliation between work and other interests or activities Accumulation of work experience
Differential advantages (institutional variables: legislation on employment contracts, collective bargaining, employment policies, labour market conditions) Lower ‘costs’ in terms of contractual and working conditions Higher earnings than unemployment benefits and better occupational prospects compared with further job search, including for disadvantaged workers who may find it difficult to find any job

Based on this brief overview of some of the main elements which may influence the diffusion of TAW, it is possible to conclude that many factors are at play and that it is therefore difficult to associate TAW with only one of the combinations cited above; rather, a combination of all of these factors is a more likely scenario.

One particular aspect worth mentioning at this point is the fact that the propensity to accept or even seek out TAW will increase, compared with the tendency to prolong job searches, if TAW can provide working conditions and career prospects comparable to other employment contracts. Such conditions could include better wages or the chance to obtain a more secure job. As a consequence, the regulation of TAW emerges as a crucial factor for balancing the demand for and offer of TAW and for promoting its diffusion as an element of an integrated labour market – as opposed to a secluded segment of a fragmented labour market. In order to overcome the risks that TAW might potentially produce in terms of the segmentation of both internal and external labour markets between core (protected) workers and peripheral (precarious) workers, the regulation of TAW should therefore aim to balance its differential advantages and to enhance its specific advantages.

Extent of TAW

Looking at international data on the diffusion of TAW, it emerges that this type of work almost invariably involves only a small proportion of the total workforce – often between 1% and slightly more than 2% of the workforce (Table 2). This seems to confirm the aforementioned finding that TAW ‘structurally’ involves a frictional, short-term element in the job-matching process. Moreover, it suggests that the advantages of TAW compared with other employment contracts or situations are not significant enough to increase the levels of demand for and offer of TAW. However, this does not mean that TAW in its current guise is beyond improvement and that its regulation is either not relevant or is fully satisfactory as it stands. The limited diffusion of TAW eventually depends on the level of demand and implies that differential advantages are not particularly relevant for companies. However, disadvantages for workers may nevertheless be substantial, particularly given the short duration of assignments and the larger number of people involved than the number of full-time equivalents considered in statistics: for instance, four people working as temporary agency workers for three months will appear in statistics as one single ‘full-time equivalent’ worker, and may in fact have been unemployed for most of the year, receiving a proportionally low income. Moreover, these data seem to point to the significant potential for the development of TAW in terms of its contribution to labour market entry, particularly for disadvantaged groups. Therefore, the scope for intervention, both by policymakers and the social partners, seems to be significant.

Another important aspect concerns the recent growth of TAW in terms of placements, full-time equivalents, the number of agencies and turnover. This clearly indicates both the vitality of the market, which is attracting an increasing number of competitors, and the growing significance and possible acceptance of such employment. Finally, it is important to draw attention to the cyclical fluctuations that follow economic downturns and upturns: in this context, companies can use TAW to accommodate the upturns in demand, possibly consolidating employment as growth is secured.

It is also important to highlight that such ‘quantitative’ homogeneity disguises a highly complex picture in terms of the institutional and socioeconomic features of TAW. In this global overview, there isn’t sufficient room to outline the national details of TAW composition or legislation. Instead, the report provides a framework which may be useful for comparing different situations and for highlighting the most crucial issues in the regulation and concrete developments of TAW. For further details and analyses, as far as European countries are concerned, readers can refer to the report entitled Temporary agency work in an enlarged European Union (582 Kb PDF) (Arrowsmith, 2006).

A number of caveats should, however, be underlined. The aforementioned quantitative marginality of TAW, as reflected by the incidence of TAW in the context of the national workforce, should not nevertheless undermine the relevance of this form of employment for a number of reasons.

  • Firstly, the data on TAW often refer to full-time equivalent workers rather than to the actual number of workers involved in this type of work. Given the short average duration of placements, the actual number of persons involved in TAW may be much higher, adding to the social importance of this type of employment contract.
  • Secondly, the data only refer to those who succeeded in finding a job through TAW. The number of people who attempted to enter the labour market through this form of employment may be significantly higher: for instance, estimates for Italy indicate failure rates of around 50% in terms of securing TAW (Micheli, 2006). Bearing in mind that TAW is considered a favourable means of employment for disadvantaged segments of the labour market, with the potential to help to foster participation and employment among ‘marginal’ segments of the workforce, it may be important for analyses to also consider the proportion of candidates who remain excluded from this particular type of employment. Unfortunately, the data included in this report is unable to address this issue at this point in time.
Table 2: Extent of TAW in the EU, US and Japan (2004)
Country Number of employees Proportion of total workforce (%) Number of companies Number of branches Turnover (€ million)
AT 44,125 1.4% 380 1,424
BE 75,131 2.2% 127 1,013 3,089
DE 399,789 1.2% 4,526 7,153
DK 6,341 0.3% 645a 440a
EL 3,503
ES 150,000 0.8% 341 1,953 2,450
FI 14,000 0.6% 400
FR 569,314 2.1% 1,000 6,299 18,400
IE 25,000 366 1,300
IT 153,000 0.6% 75 2,400 4,000
LU 7,135 1.6% 40 150
NL 157,000 2.5% 1,250 4,200 6,500
NO 22,784 1.0% 632 654
PT 45,000 0.9% 247 650
SE 35,000 1.0% 550 929
UK 600,000b 1,434,098c 2.6%b 5.1%c 6,500 10,000 34,693
US 1,200,000d 2,900,000e 0.9%d 2.0%e 7,000 30,000
JP 890,000 3.4% 20,278 17,150

Notes: (a) 2000; (b) Department of Trade and Industry (DTI) estimate (2003) reported by the Trades Union Congress (TUC); (c) Employer organisation (REC); (d) US Government Accountability Office (GAO), 2005; (e) American Staffing Association (ASA), 2005

Source: EIRO national centres, various years

Legal framework

In the EU, TAW is now an established component of the labour market, developing more rapidly in recent years. However, the legal regulation of TAW is only a relatively recent phenomenon in many countries, dating back to the 1960s and 1970s in only about half of the original 15 EU Member States (EU15). A majority of countries made this type of employment contract legal in the late 1980s and 1990s or even later, including many of the 12 new Member States (NMS) which joined the EU in 2004 and 2007 and where TAW is reportedly still ‘underdeveloped’. In fact, in some of these countries – such as Bulgaria, Cyprus, Estonia, Latvia, Lithuania and Malta – little or no statutory regulation of TAW existed in 2005, when the aforementioned report was drafted (see Arrowsmith, 2006). A similar situation regarding the late regulation of TAW can be found in Japan, where this form of employment was first introduced in 1986. Among the advanced economies covered in this report, only the US reports the presence of TAW since the 1930s. However, the typical triangular arrangement of TAW, with the agency as the employer, started to emerge as the standard formal employment relationship only in the 1950s in the US, following a decision by the Internal Revenue Service (IRS) to grant employer status to the agencies for the purposes of collecting taxes, eventually becoming the general rule by 1971. By this time, every US state had recognised temporary work agencies as the legal employer of temporary agency workers, instead of the user companies, as was formerly the case.

The recent history of TAW suggests that this form of employment can be included among the innovations in labour policies which were introduced to cope with the high unemployment levels that followed the economic downturn of the 1970s and 1980s, with a view to increasing labour market efficiency. In fact, such a description has become even more evident in recent years, with the modification and reform of many of the rules that were originally introduced.

In the EU in particular, but also in Japan, a progressive easing of the strict regulations that were typical of the first experiences of TAW can be observed. The main reasoning behind the introduction of regulations, and notably restrictions on the use of TAW, was to prevent abuses and an evasion of the labour laws. On the one hand, the introduction of TAW was linked to the fight against irregular work and to uncovering the use of irregular work in companies; on the other hand, restrictions were meant to limit TAW to actual short-term adjustments in the company workforce, so that TAW would add to existing employment rather than act as a substitute for it. Therefore, the prevailing attitude in the earliest interventions could be described as one of ‘limited consent’. As a result, the provisions usually included an explicit definition of the reasons for using TAW, of restrictions on the duration of placements, and of limitations placed on the sector and on occupations involved in this type of employment relationship.

In recent times, a more liberal stance has emerged with respect to TAW, with certain ‘liberalisation’ measures being put in place in the great majority of countries, thus opening up the scope for TAW. At the same time, greater efforts have been made in the area of temporary workers’ rights, with a view to establishing stricter conditions in relation to equal treatment and to granting increased protection.

Examples include the Netherlands, the first EU country to introduce TAW in 1965. In 1998, the country abolished the licensing system and a number of restrictions relating to placement, maximum duration and worker redeployment. In 1999, the Netherlands went a step further towards the ‘liberalisation’ of TAW, following an agreement between the social partners. At the same time, the Flexibility and Security Act (Wet Flexibiliteit en Zekerheid) defined the legal position of temporary employees as a standard labour contract between the temporary employee and the temporary work agency, granting participation rights to temporary agency workers in the user enterprise. This act also confirmed other pre-existing protections, such as an equal wage clause for temporary agency workers.

In 2002, Germany enacted a major revision of the 1972 Act on Temporary Employment Business (Arbeitnehmerüberlassungsgesetz, AÜG); this revision lifted many of the restrictions, while stipulating that temporary work agencies have to guarantee equal pay and employment conditions comparable to those granted to permanent staff in the user company. However, collective agreements can provide for exceptions to this principle. Similar interventions aimed at relaxing the conditions on the use of TAW while reinforcing the rights of temporary agency workers took place in Austria, Portugal and the UK.

The UK represents a particularly interesting case, since its legislative framework can probably be deemed the most liberal across Europe. In 2003, a revision of the Conduct of Employment Agencies and Employment Businesses Regulations was introduced to simplify the requirements for TAW, thereby increasing flexibility and reducing costs for employers. However, this revision also improved the rights of temporary agency workers by extending restrictions on fees, making temporary work agencies responsible for employees’ health and safety in the user company and for informing workers about the terms and conditions of their employment.

In many EU countries, such as France Italy and Norway, recent reforms have expanded the scope of TAW. A similar shift has occurred in Japan, where limitations on the length of placements as well as on the sectors and occupations covered by TAW have been progressively reduced. Among the most notable changes in this respect is a 2004 revision of the legal framework, which also allowed for TAW in manufacturing.

Thus, a new balance is emerging between the regulation of the TAW market and the protection of temporary agency workers, with less restrictions being imposed on the former and more interventions being introduced for the latter. Many of the countries have therefore shifted from an ‘input-centred’ regulation framework to a mainly ‘outcome-centred’ one, which relies more on the market and seems to recognise the positive contribution that TAW can make to an efficient labour market.

This latter aspect is particularly evident in countries where the potential role of TAW has been recognised by the inclusion of temporary work agencies in the system of employment services; this is especially apparent with respect to measures devised for the job placement of disadvantaged workers through specific incentives and rules. In Belgium, where regulation has remained otherwise quite strict, a reform introduced in 2000 has offered temporary work agencies a role in the placement of specific segments of the labour market, including people who are long-term unemployed, provided that these people are hired on open-ended contracts. Similarly, in Germany, new legislation in 2002 assigned an important role to temporary work agencies in the placement of unemployed people, even those on short-term assignments, with a view to supporting their progressive and full reintegration in the labour market. In France and Italy, recent legislation – namely the Social Cohesion Framework Act of January 2005 and the labour market reform of September 2003 – has substantially increased the role of temporary work agencies in job placement services. In Japan, the possibility of using TAW as a form of job placement was introduced in 2002 through the contract ‘for prospective placement’, whereby the user company pledges to directly hire the worker at the end of the assignment.

Current regulations

Thus, the only area of increasing regulation in TAW appears to be in relation to the enforcement of equal treatment and the introduction of specific support measures for temporary agency workers belonging to particular disadvantaged groups; conversely, other areas of regulation are gradually being abandoned (see Arrowsmith, 2006). In 2005, it was found that Belgium and Portugal had the strictest regulations among the EU countries in relation to TAW; in these two countries, the equal treatment principle, as well as restrictions on reasons for use, duration and sectoral and/or occupational coverage, remained in force. Among the NMS, countries such as Poland and Romania appeared to have the most stringent regulations. Conversely, Denmark, Ireland, Norway, Sweden and the UK had the least restrictive regulations, not even providing for equal treatment provisions.

Outside of Europe, the US TAW market appears to be particularly unregulated. For instance, no statutory definition of TAW exists. Only in four states – namely, Massachusetts, North Carolina, New Jersey and Rhode Island – does a specific regulation exist in this respect; however, this regulation does not provide for any operational constraints, but only stipulates licensing, registration and certain other requirements regarding the business entity. Furthermore, no equal treatment principle applies in the US, and the wages of temporary agency workers are often substantially lower than those of permanent workers, averaging at around 20% less in most occupations. Some recent attempts made at federal and state levels to regulate TAW have failed – more specifically, in relation to enforcing equal treatment in pay and other dimensions, such as health and safety protection. In Rhode Island and South Carolina, legislation on the ‘right to know’ was passed, whereby agencies must provide temporary agency workers with job descriptions, pay rates and work schedules. It is also worth mentioning that in the US, it is possible to replace striking workers and that, in some cases, temporary agency workers have been used for this purpose.

This lack of statutory protection in the US has sometimes prompted legal action challenging discrimination. A groundbreaking case in this respect concerns the ‘Vizcaino versus Microsoft’ case. In this instance, Microsoft was obliged to indemnify temporary agency workers who had been interviewed and selected directly by Microsoft, but who had been classified as temporary or freelance workers (Van Jaarsveld, 2004). In doing so, the company had avoided granting workers a series of benefits, including stock options. The court decision sanctioned the fictitious separation between the company workers and the temporary agency workers.

Meanwhile, in Japan, where TAW has been widely regulated, restrictions have been relaxed somewhat in recent years, particularly with regard to sectors and professional categories, as well as the duration of assignments. Some limited restrictions are still nevertheless in place in these respects.

Main sectors, occupations and workers

The widespread use of TAW across sectors and occupations derives from its potential to contribute to the adaptability of companies and to be used in different contexts. As the various limitations which had constrained the use of TAW in the past are being increasingly eliminated, it can be expected that the use of TAW – principally, the demand for TAW – mostly depends on the specific advantages of TAW in the various sectors and occupations. Therefore, a disproportionate distribution of TAW across sectors and occupations would suggest that the adaptability offered by TAW is particularly valuable for the sectors and occupations where it is most prevalent.

In terms of the sectors involved in TAW, data available for 13 of the EU Member States and Norway identify three basic groups of countries: the first group shows a concentration of TAW in manufacturing (Austria, France, Hungary, the Netherlands, Portugal and Slovakia); the second group reveals a concentration of TAW in the services sector (Spain, Sweden and the UK); the third group shows a more balanced situation (Belgium, Denmark, Finland and Italy). Of course, this situation may vary over time. For instance, in Italy, TAW has remained almost exclusively concentrated in manufacturing and metalworking for a long time; however, in recent years, a more balanced distribution of TAW across sectors has been observed. In the US, TAW is mainly found in manufacturing and business services, which together account for nearly 60% of all temporary agency workers. In Japan, the distribution of TAW across sectors is more balanced: around 35% of all temporary agency workers are employed in manufacturing and construction, while similar proportions can be found in the three service sectors benefiting the most from this form of employment – namely, transport, financial services, and wholesale and retail. The remaining proportion of TAW is scattered across the country’s other service sectors – a picture which roughly corresponds to the distribution of work across the entire economy.

In relation to the distribution of TAW across occupations in Europe, ‘a significant proportion of TAW appears to be located in lower-skilled work in the service sector or manufacturing, and in clerical and administrative occupations’ (Arrowsmith, 2006, p. 8). Although detailed data on the types of occupations involved in TAW are not available for Japan, it is possible to conclude that a similar picture also holds true for this country, where the majority of temporary agency workers are employed in administrative tasks, such as the operation of office equipment, filing and general office work. The fact that a quarter of all workers in Japan have a university degree or a higher education suggests that a significant proportion of temporary agency workers may be overqualified, even if this is mainly the effect of the significant presence of young workers entering the labour market. In the US, the overall distribution of TAW across occupations, according to the ASA (the association of temporary work agencies), shows an almost equal presence of lower-to-medium white-collar and blue-collar occupations (clerical and industrial jobs) and medium-to-higher occupations (professional-managerial, technical, information technology, healthcare), even if other analyses indicate that higher-skilled occupations represent only about 10% of all TAW.

Analysis of the demographic characteristics of temporary agency workers in Europe, the US and Japan shows that the gender balance is clearly influenced by the sectoral distribution: in instances where manufacturing prevails in TAW, men represent the majority of the workers involved; in the other situations, men and women are almost equally represented among temporary agency workers, or women slightly prevail, as is evident in the US and Japan. Another important aspect relates to the overrepresentation of young workers, a fact which seems to suggest that TAW often acts as an entry point for workers into the labour market.

Employment conditions

The first important element to be taken into consideration in relation to employment conditions is identifying the legal employer. From a legal point of view, it is worth mentioning that only two of the countries covered in this report do not consider TAW as a triangular employment contract with the temporary work agency as the employer – namely, the UK and Ireland. In the UK, most temporary agency workers are employed on a service contract; as a result, there is no proper ‘employer’ as such. In Ireland, temporary agency workers are considered to be employed by the user company – a step taken to extend the coverage of rules on unfair dismissal to temporary agency workers.

Duration of assignments

Another important element relates to whether employment contracts are of indefinite duration or temporary. Although it is usually possible for temporary work agencies to hire workers on both open-ended and fixed-term contracts, the former are quite rare while the latter represent the standard arrangement, as workers are usually hired to cover individual assignments. Exceptions to this trend in Europe are the countries Sweden, Germany and Hungary. In Sweden, temporary agency workers are in fact considered as employees with permanent contracts, while in Germany, it is normal practice to employ temporary agency workers on open-ended contracts; in Hungary, some 60% of TAW is on the basis of permanent contracts. In Japan, so called ‘regular’ dispatched workers – that is, those employed on contracts of indefinite duration by the temporary work agency – represent some 20% of all workers. In the US, temporary agency workers are usually hired for the duration of the assignment in question.

The duration of TAW assignments is generally rather short, often less than a month and seldom longer than six months. In fact, in only a handful of the countries considered do a significant proportion of TAW assignments last at least six months. In Austria, more than 60% of all white-collar temporary agency workers have an employment contract lasting at least six months, although this percentage drops to 28% for blue-collar workers. Other countries in which there is a significant level of longer TAW assignments include the UK (45% of all contracts last at least six months), Belgium (25% at least six months) and Ireland (15% at least five months); moreover, in Portugal and the Netherlands, the average duration of assignments is as long as four and five months, respectively.

In the US, TAW assignments last, on average, about three months. However, this finding encompasses two quite distinct phenomena: the use of TAW to fill permanent jobs, thus avoiding the costs entailed by hiring employees directly on open-ended contracts – in the US, these workers are often referred to as ‘permatemps’; and giving workers very short assignments lasting only a day or even a few hours. In Japan, the proportion of temporary agency workers with an employment contract of at least six months is quite significant, reaching over 35% of these workers. Moreover, about one fifth of all temporary agency workers in Japan are employed on a contract of an indefinite duration. Possibly, these may be ‘regular’ temporary agency workers who are sent to user companies for an indefinite period.

The length of assignments constitutes a crucial element in assessing the capacity of workers to earn a living through TAW. Furthermore, despite the formal parity between temporary agency workers and employees in the majority of European countries, the length of assignments may bear an influence on the possibility of accessing various benefits attached to employment – both at company level, where a number of benefits may be linked to length of service, and at national level, since certain allowances, such as unemployment benefits, are linked to the length and continuity of previous employment. A proper analysis of this source of vulnerability would require data on the duration of employment of individual workers, rather than of individual assignments, in order to take into account the continuity ensured by subsequent assignments; information would also need to be obtained on the proportion of temporary agency workers who make a transition into permanent employment through TAW. Unfortunately, this information is not readily available. However, the combination of data on the length of assignments, the high turnover which characterises TAW, and the fact that only a minority of temporary agency workers – largely those with a stronger position in the labour market – make the transition into permanent employment, suggest that this represents an important area for intervention.

This can be illustrated by the case of Japan, where equal treatment exists at a substantial level and where temporary agency workers do not have particularly lower hourly wages compared with other workers. Nevertheless, an estimated pay gap as significant as 30% is still evident among this group of workers (Bank of Japan). According to a survey by Japan’s Ministry of Health, Labour and Welfare, many temporary agency workers do not receive paid holidays (for which at least six months’ continuous service is required), allowances for commuting and housing, nor company bonuses.

A more substantial difference between temporary agency workers and permanent employees can be found in the US, where no equal treatment provision is granted under the institutional framework for TAW. General contracts between temporary agency workers and temporary work agencies stipulate that the former are ‘not entitled to holidays, vacations, disability, insurance, pensions or retirement plans, or any other benefits offered or provided by [a given company] to its direct employees’. Thus, only a limited proportion of temporary agency workers in the US receive such benefits from their agencies: between 9% and 14% of temporary agency workers are covered by health insurance; some 6% receive life insurance; a further 6% are included in ‘401k retirement plans’ (a retirement savings plan which is funded by employee contributions and often matching contributions from the employer). In terms of pay, temporary agency workers earn, on average, 20% less than those who are directly employed by the company. Moreover, the limited duration of TAW assignments makes temporary agency workers three times more likely to have an annual income below USD 15,000 (about €10,909 as at 8 August 2007), compared with standard full-time workers with comparable socioeconomic characteristics. Furthermore, it also decreases the possibility that temporary agency workers can benefit from certain provisions, such as the Family and Medical Leave Act (FMLA), which entitles workers to unpaid leave of up to 12 months to care for an infant or a sick family member.

Social partners and collective bargaining

The specificities of TAW, in particular the triangular employment relationship, make collective relations in this sector more complex compared with standard sectors. Firstly, a substantial asymmetry often exists between the employer’s side and the worker’s side. While employer organisations are usually present and show high membership rates, trade union density is often quite low in the TAW sector. The short duration of assignments, and even more importantly the short time spent in employment during a given year, along with the possible changes occurring in sectors and job positions, contribute to making trade union membership less likely. Only in Italy do specific trade union associations exist for temporary agency workers; these unions usually bargain at sectoral level under the umbrella of trade union confederations.

TAW can be considered a sector of its own as far as the labour market is concerned. Nevertheless, basic employment and working conditions, including pay, are substantially determined by the user companies belonging to the different sectors. Ironically, this variability increases when workers are protected against discrimination by equal treatment clauses, as the traditional regulation of the employment relationship tends to fall outside the regulatory capacity of the two parties of the industry. In practice, trade union representation and sectoral collective bargaining essentially concern temporary work agency employees or temporary agency workers on open-ended contracts, during the periods when they are not assigned to user companies. Therefore, collective bargaining in TAW tends to address new issues that pertain specifically to the relationship between the agency and temporary agency workers and which concern issues such as skills development and measures aimed at retaining workers. It is interesting to observe that in this case, the interests of the agencies – in particular, strengthening their competitiveness in terms of the skills they can provide to user companies – are capable of meeting the workers’ demand for more stable employment and increased employability.

This convergence, and specifically the interest of temporary work agencies in developing a regulatory framework which improves their competitive edge, may help to explain the apparent contradiction regarding the significant development and crucial role of collective bargaining in Europe, mostly at sectoral level, despite a relatively weak trade union presence in the sector. The NMS represent a notable exception in these respects, as 2005 data reveal that sectoral collective bargaining was absent in all 12 countries, while no association representing temporary work agencies had been set up in over half of these countries.

The clearest example of the regulatory role of collective bargaining in the TAW sector can be found in the Netherlands, where a sectoral agreement includes provisions for training and explicitly links the duration of employment with the agency to the type of employment contracts, irrespective of the number of assignments. Accordingly, after 78 weeks’ work with the temporary work agency, the temporary agency worker will be hired on a fixed-term contract lasting two years; this means that the agency will have to find suitable jobs until the worker’s contract expires and that the temporary agency worker will be paid during idle periods between assignments. After 3.5 years with the same agency, the temporary agency worker is entitled to an open-ended contract with that agency. Other examples of similar interventions include sectoral collective agreements in France, which have led to the introduction of the Temporary Work Social Action Fund (Fonds d’Action Sociale du Travail Temporaire, FAS-TT) and the Temporary Work Training Insurance Fund (Fonds d’Assurance Formation du Travail Temporaire, FAF-TT); the FAS-TT provides support in areas such as housing, consumer credit and study grants, while the FAF-TT offers assistance in training and study leave. In Italy, sectoral bargaining has led to the establishment of ad hoc bodies providing services for temporary agency workers, and notably training services through a fund created by legislation.

Collective bargaining plays a crucial role in Denmark and Sweden, where no statutory regulation of TAW is present. It is important to note that collective bargaining in Sweden follows occupational lines and that a number of collective agreements therefore exist – namely, agreements for office and commercial workers, graduate engineers, white-collar and professional workers, and blue-collar workers. In Sweden, two provisions are particularly noteworthy: the stipulation that temporary agency workers should be hired on open-ended contracts, as already mentioned, and an income guarantee of up to 80% of the normal monthly wage for white-collar and professional workers, even if no gainful work is performed. In Denmark, multiple collective agreements also exist, depending on the different sectors, such as education, building and construction, storage and transport. It should be highlighted that the multiplicity of agreements along occupational or sectoral lines overcomes the difficulty of low trade union density among temporary agency workers and those connected to the multi-sectoral nature of TAW. Separate collective bargaining for different occupational groups involved in TAW is also present in Norway.

In Japan, there is no sectoral collective bargaining, as industrial relations are generally performed at company level. Often, temporary agency workers do not have access to company trade unions, since they only represent permanent staff; this greatly hinders the capacity of temporary agency workers to benefit from collective representation. Some examples do exist of ‘regular’ temporary agency workers organising a trade union within the company they work for. However, the impact of such efforts appears to be marginal. According to a survey on trade union activities conducted by the Ministry of Health, Labour and Welfare, while almost 70% of company or establishment trade unions in Japan declared that temporary agency workers were present in their workplace, only 14.9% of them carried out some kind of activity favouring temporary agency workers, while just 3% made demands to improve the working conditions and treatment of temporary agency workers.

In the US, trade unions have attempted to limit the use of TAW through collective bargaining, in order to avoid the practice of hiring ‘permatemps’. In some cases, collective agreements have stipulated the maximum number of days, weeks or months that a temporary agency worker can be employed for within a particular company, sometimes requiring that laid-off trade union members should be recalled to fill temporary positions rather than a temporary agency worker. Other provisions include limiting the proportion of temporary agency workers to a certain percentage of the workforce. Some agreements, as in the case of negotiations between the Office and Professional Employees International Union (OPEIU) and the non-profit organisation Kaiser Permanente, have even introduced an equal hourly pay rate for temporary agency workers and regular employees. However, these examples concern only a limited proportion of temporary agency workers, the majority of whom remain largely excluded from collective bargaining coverage.

TAW in Brazil, China and India

As illustrated, the diffusion and characteristics of TAW largely depend on the institutional framework, as well as on the ways in which it can meet the needs and demands of both companies and workers. The combination of these factors takes a relatively specific shape in the emerging countries covered by this report – namely, Brazil, China and India.

In Brazil the regulation of TAW is quite extensive. Firstly, temporary work agencies can only operate in urban areas. As in many other countries, agencies need to be registered at the country’s Department of Labour and must meet a minimum capital requirement. TAW can be used to meet the transitory and additional needs of companies. Moreover, temporary agency workers are considered equal to permanent employees. Data on both fixed-term work and TAW in Brazil show that they each represent no more than 2% of total formal employment. In addition, temporary jobs are more likely to involve young workers.

In China, TAW, as it is understood in this report, does not exist. Nevertheless, there is evidence of daily labourers who are placed by ordinary job centres, mainly to carry out low-skilled tasks in families; such tasks include home removal, domestic maintenance and decoration services. The workers involved in these areas are mainly young men who work according to verbal agreements. This phenomenon emerged in the late 1990s, although it has increased very little over time. Such work constitutes only a marginal aspect of the Chinese labour market and involves a limited number of persons, along with a substantial lack of any statutory regulation; as a consequence, it is not always clear who should be regarded as the employer. The issues raised by this type of employment have so far failed to attract the attention of the social partners, and there are no signs of movement towards a formal regulation of TAW in China.

In India, although there is evidence of TAW, it is only present to a limited extent. Due to the presence of a large informal sector, formal temporary agency workers usually enjoy better economic and working conditions than average; this may also be related to the fact that the user companies tend to be large firms. Moreover, some temporary work agencies are emerging as major employers and are starting to provide their employees with welfare protection schemes and training. Overall, therefore, the extent of TAW appears to be increasing in India and there are indications that its relevance may grow in the near future.

Small and medium-sized enterprises

Industrial relations in SMEs have often attracted the attention of researchers and practitioners, particularly since restructuring and reorganisation processes have led to widespread downsizing of companies and the outsourcing of services. These developments have clearly marked the decline of large, vertically-integrated organisations to the benefit of more dynamic, smaller enterprises. The changes in labour relations brought about by this shift have in turn generated increased interest in SMEs.

Large companies in the manufacturing sector, as well as large bureaucracies in public administration, have long been the strongholds of trade unions and industrial relations, as well as the main environment for innovation in human resource management (HRM) practices. In fact, large manufacturing companies were among the first places where trade unions and collective bargaining emerged and consolidated; it is in such companies that the collective identities, which have fuelled industrial relations developments throughout the 19th century, were shaped by daily work experience and strengthened by the ideology of the working class movements. Indeed, studies and research, as well as theoretical analyses, almost exclusively focused on the features and consequences of industrial relations within large organisations. At the same time, industrial relations were driven by mobilisations in large companies and often mainly confined to large workplaces. As a consequence, SMEs received limited attention, if any, and were mainly considered as representing traditional, sometimes backward, work environments.

During the early phases of industrial restructuring, in the 1970s and 1980s, downsizing and outsourcing activities, and the ‘disintegration’ of the production process, which increasingly led to the inclusion of smaller companies in production chains, were considered to imply a worsening of working conditions and the erosion of collective rights. Such negative consequences may have been linked to changes in the applicable collective agreements, or even to the absence of any collective bargaining coverage among the subcontractors; at the same time, they may have been attributed to a weaker trade union presence and the greater difficulties involved in enforcing worker and trade union rights, if not the labour regulations as a whole. Even today, many trade union and worker representatives still maintain a negative attitude towards the outsourcing of services. Such a position appears to be based on the implicit or explicit assumption that reorganisation, and outsourcing in particular, are always aimed at reducing production costs directly, including labour costs, and that this can only be achieved at the expense of either working conditions or workers’ rights, or both.

However, such a view may be somewhat biased. Firstly, the objectives of reorganisation processes are broader than simple cost reduction and may include other goals, such as upgrading skills, delegation and cooperation, which may have a positive impact on working conditions and worker involvement. Secondly, outsourcing and subcontracting increasingly involve large specialised companies, sometimes multinationals, where industrial relations and HRM practices can be strong and well-developed. Thirdly, SMEs are now seen as an important component of competitive systems, both for employment creation and in terms of the crucial role that these companies can play in innovation processes.

SMEs make up quite a fragmented picture, where it is possible to find examples of both traditional and progressive workplaces, paternalistic and innovative HRM practices, measures to avoid trade unionism and fully-fledged industrial relations systems. For this reason, presenting a general view of the ‘average’ SME can be difficult and misleading. Rather, it is more beneficial to identify the main distinctions that can be drawn within the very broad spectrum of SMEs. However, this represents a difficult task, given the lack of information and the persistent knowledge deficit in this area, despite a number of recent initiatives which have attempted to fill this gap.

A further complication is the variety of national situations. The definition of a SME can vary significantly across countries, as reflected by the different definitions adopted in each national context. The variation in the distribution and relevance of SMEs in the economy and across sectors may also be significant, as is the size of the informal sector, which is essentially made up of micro and small enterprises. This latter aspect is of particular importance in relation to labour and industrial relations, since, almost by definition, the informal sector is characterised by irregular jobs, lack of enforcement of regulations and, as a result, poor employment and working conditions.

Cross-country differences in defining SMEs

An initial insight into the differences that exist across countries in relation to SMEs can be provided by looking at the various definitions adopted by the countries. While such differences make comparisons more difficult, they can also provide some clues regarding the characteristics of the various national production systems: accordingly, the higher the threshold for SMEs (number of employees), the greater the potential relevance of large companies for the national economy. Therefore, the variety of definitions may give an insight into the ‘quality’ of the SMEs’ role in a national context.

At EU level, the increasingly common ceiling used to define SMEs is 249 employees: this encompasses a further distinction between micro-enterprises (less than 10 employees), small-sized enterprises (10–49 employees) and medium-sized enterprises (50–249 employees). However, the definitions adopted at national level vary markedly. For instance, in Germany, medium-sized enterprises consist of those with up to 499 employees, whereas the corresponding threshold in Spain is 200 employees, while in Denmark it is 100 employees. This finding is a clear indication that the production systems of these economies differ substantially: in Germany, the economy is more centred on large companies, while in Spain and Denmark in particular, it is more focused on smaller enterprises. Similarly, the US definition of SMEs, which is based on a threshold of 500 employees, reflects the fact that the US economy has traditionally been dominated by large companies.

In Japan, a more comprehensive approach has been adopted in this respect, which differentiates between sectors and takes into consideration employment levels and registered capital. Accordingly, in line with the Small and Medium Enterprise Basic Law, the upper limit for SMEs is as follows: 300 employees or JPY 300 million (about €1.8 million) of registered capital in manufacturing; 100 employees or JPY 100 million (around €600,000) in wholesale; 50 employees or JPY 50 million (about €300,000) in retail; and 100 employees or JPY 50 million in services. Micro-enterprises are defined according to employment levels only, as follows: up to 20 employees in manufacturing; and up to five employees in wholesale, retail and services in general. In this case, the different diffusion of SMEs in the various sectors has been taken into account, in order to provide for a more focused distinction.

In China, SMEs are defined on grounds of sales turnover, and fixed assets in the case of small companies, rather than on the basis of the number of employees. Small enterprises are defined as those with up to CNY 50 million (about €4.8 million) in sales volume or CNY 1.8 million (around €175,000) in book-value fixed assets; medium-sized companies are defined as those with annual sales of between CNY 50 million and CNY 500 million (about €48 million). Turning to the employment size of these companies, the industrial census shows that the average size of small enterprises in China is about 15 employees, while in medium-sized enterprises, the threshold increases to 893 employees. In Brazil, the definition of SMEs includes all establishments with up to 249 employees. In India, the so-called ‘unorganised sector’ consists of informal activities and all unincorporated private enterprises owned by individuals or households with less than 10 workers.

Beyond the specificities of these national definitions, and the information which can be derived from such definitions, it is important to consider from a substantive point of view whether the absolute number of employees in itself provides a clear indication of specific features, problems or challenges that are likely to concern all enterprises within a certain category. If the latter is true, a common definition would remain meaningful across countries. Otherwise, a relative rather than an absolute definition of SMEs would be preferable. In general, it is possible to say that the number of employees can provide a good indication of the types of problems and issues that each company is likely to face.

Role of SMEs in economy

Despite the differences in the definitions and characteristics of SMEs, which are evident in the countries under consideration, SMEs demonstrate a number of common features across the countries:

  • SMEs are by far the most prevalent organisational form, amounting to between 98% and 99% of all enterprises;
  • SMEs employ at least 50% of all workers;
  • their number and weight in terms of employment has generally increased in recent years.

Of course, these general trends may hide important differences. For instance, a decline in SMEs in certain sectors may be balanced out by an increase in other sectors. Diverging patterns are quite common within countries and also across countries (Table 3).

Table 3: Profile of SME sector in EU, Japan and US
Country No. of companies (% and thousands) No. of persons employed (% and thousands)
  Micro Small Med- ium SMEs Large Total (thous-ands) Micro Small Med-ium SMEs Large Total (thous-ands)
AT 86.9% 11.2% 1.9% 99.6% 0.4% 268 37.2% 18.9% 15.7% 71.9% 28.1% 2,938
BE 93.2% 5.7% 0.9% 99.8% 0.2% 438 4.0% 16.4% 12.9% 69.5% 30.5% 3,264
DE 88.0% 10.2% 1.5% 99.6% 0.4% 3,019 34.0% 18.0% 12.8% 64.8% 35.2% 30,884
DK 87.4% 10.2% 1.9% 99.5% 0.5% 206 35.7% 20.0% 16.9% 72.6% 27.4% 2,082
EL 97.5% 2.1% 0.3% 100% 0.0% 771 56.8% 16.9% 12.9% 86.6% 13.4% 1,785
ES 93.4% 5.8% 0.7% 99.9% 0.1% 2,677 50.5% 19.6% 11.5% 81.7% 18.3% 15,637
FI 93.2% 5.4% 0.9% 99.5% 0.5% 222 34.5% 15.2% 14.8% 64.5% 35.5% 1,523
FR 93.0% 5.8% 1.0% 99.8% 0.2% 2,501 37.1% 16.1% 1.3% 66.6% 33.4% 18,932
IE 85.6% 12.4% 2.1% 100% 0.0% 97 25.2% 23.5% 21.1% 69.8% 30.2% 929
IT 95.6% 3.9% 0.4% 99.9% 0.1% 4,489 56.8% 16.9% 9.8% 83.5% 16.5% 18,834
LU 87.5% 12.5% 4.2% 100% 0.0% 24 24.4% 24.4% 24.4% 73.3% 27.1% 225
NL 90.4% 7.5% 1.6% 99.7% 0.5% 572 31.7% 17.9% 15.6% 65.2% 34.8% 6,951
PT 93.4% 5.6% 0.9% 99.9% 0.1% 694 374.1% 231.5% 183.5% 789.1% 211.2% 321
SE 93.4% 5.6% 0.8% 99.8% 0.2% 486 3.8% 16.0% 13.6% 68.0% 32.1% 3,228
UK 89.3% 9.0% 1.4% 99.6% 0.4% 2,234 32.1% 14.8% 12.3% 59.2% 40.8% 247,30
EU15 92.4% 6.5% 0.9% 99.8% 0.2% 18,698 39.7% 1.7% 12.8% 69.7% 30.3% 135,152
JP* 72.2% 22.2% 4.9% 99.3% 0.7% 1,530 11.0% 21.6% 25.0% 57.7% 42.3% 31,562
US** 78.5% 19.7% 1.5% 99.7% 0.3% 5,767 11.0% 25.1% 14.5% 50.7% 49.3% 113,398

* In Japan, medium-sized enterprises include those with up to 299 employees. Data refer to 2004.

** In the US, small-sized enterprises include those with up to 99 employees, whereas medium-sized businesses have up to 499 employees.

Source: Observatory of European SMEs (2003) for EU15; Statistics Bureau and the Ministry of Internal Affairs and Communications for Japan; the Small Business Administration (SBA) for the US

An important issue concerns the possible impact of globalisation on SMEs, which is generally thought to have brought increased challenges and risks, as well as more opportunities. In particular, the difficulties that some traditional manufacturing sectors are facing in advanced economies have probably affected SMEs to a greater extent, given the limitations in resources that are available to SMEs to face restructuring. This is evident, for example, in the case of industries involved in the manufacturing of textiles, clothes and shoes, as witnessed by the decline of some clusters in Italy. Of course, globalisation may also correspond to the strengthening of competitors in emerging countries, which may easily include SMEs. Moreover, in some cases, SMEs in advanced economies are delocalising to other countries, thereby contributing to the ‘transfer’ of SMEs from one country to another.

Another factor is the possible impact that the restructuring of large enterprises may have on the supply chain of SMEs. In Japan, for instance, SMEs have been markedly affected by the reorganisation and delocalisation of larger enterprises. Offshoring has entailed, in this case, closures and the delocalisation of SMEs which have followed the larger companies; at the same time, it has involved the redefinition of SMEs’ positioning in the relative market, with companies extending their customer portfolio to increase their client base, thereby abandoning a longstanding Japanese practice of (almost) exclusive subcontracting relationships.

Working conditions

Data on working conditions in SMEs are still limited due to the scarcity of available sources of information. An initial problem in the study of SMEs relates to their uneven distribution across sectors. Some of the sectors in which SMEs prevail in terms of employment are characterised by low levels of education and skills, low value-added production and low productivity; thus, it is not surprising to find less favourable working conditions in such sectors, especially in relation to pay. In order to obtain a correct insight into the impact of company size alone on working conditions, data which would at least distinguish between SMEs and larger enterprises by sector would need to be obtained. Such an analysis has rarely been conducted and, in many cases, it is not even possible to distinguish between the different sized enterprises.

A general comparison between SMEs and larger enterprises can be misleading; moreover, comparing sectors in which SMEs are particularly widespread to national averages or to sectors in which SMEs are rare may be even less reliable. At best, both provide only weak indications of the influence of size alone on working conditions. It is not possible in this report to identify any causal links between size and sectoral working conditions, and questions arise as to whether such conditions should be regarded as sector-specific or size-specific. Thus, this report is not able to disentangle sector and size effects, which could reinforce or counterbalance each other.

Where it is possible to assess these two effects, the relevance of the direct impact of size can be somehow reduced, despite its significance. For instance, in Germany, the average pay level in the hotels and catering sector amounts to 70% of the national average wage. However, looking at the sectoral data, it is possible to conclude that this gap relates more to the sector than to the size of the company, since the average pay level of SMEs within the hotels and catering sector amounts to around 90% of the average pay levels of larger companies in the same sector.

Many SMEs operate in sectors characterised by low pay and working conditions; in certain circumstances, it can be expected that the potential negative impacts of sector and size may further worsen the worker’s situation, as could occur, for instance, in the case of a micro informal enterprise involved in the fabric dyeing sector. However, looking at other sectors in which SMEs are prominent, such as consultancy or software development, it is possible to find far more favourable working conditions, at least in relation to certain dimensions such as pay and possibly autonomy, if not working hours and employee workload.

Another limitation of the available data is that, despite all of the qualifications given earlier regarding the differences among SMEs, analysis of overall situations is usually required. Wherever possible, however, this report will try to complement such general indications with more specific references to certain categories of SMEs.


Looking at the data on working conditions, it is possible to conclude that the traditional view is generally confirmed: SMEs tend to be associated with poorer working conditions in virtually all dimensions. For instance, where available, data on pay in European countries reveal that a pay gap of between 20% and 30% compared with the national average is quite common.

Similar findings can be observed in the US: for example, in a comparison between companies with over 100 workers and those with less than 25 workers, managers and entry-level workers in the larger companies benefited from higher wages of 27% and 17%, respectively. Interestingly, US data show that the pay gap between SMEs and larger companies is higher for white-collar workers than it is for blue-collar workers. Moreover, these differences seem to increase the higher the wage, suggesting a relative disadvantage also in terms of prospects for career development in SMEs; this may be connected to the size of the company and the lack of higher positions. In the US, the pay gap is compounded by the lower access of SME employees to healthcare coverage and pension plans provided by companies. Only 60% of workplaces with up to 99 employees offer healthcare coverage, while over 95% of companies with 100 employees offer such benefits. Similarly, retirement plans are offered in less than half of small establishments, as defined earlier, compared with 90% of larger companies.

In Japan, the difference in terms of monthly total earnings is even higher. Across all sectors, micro-companies (5–29 employees) grant wages lower than 50% of those afforded by the largest companies (1,000 employees or more), while the pay gap compared with the national average wage is 20%. In fact, in Japan, the impact of both the sector and size of the company seems to be relatively high. For instance, in the hotels and catering sector, the overall monthly wage amounts to only 42% of the national average; at the same time, employees in smaller establishments receive only about half the wage of their fellow workers in larger companies. While the average pay levels in workplaces with 100 employees or more range between 75% and 82% of the national average, they range between 46% and 53% of the national average in establishments with less than 100 employees. This implies that pay levels rise as the size of the company increases. This general finding seems to largely apply to all national contexts: on average, the smaller the size of the company, the lower the pay levels.

When SMEs operate in or on the fringes of the informal sector, the pay gap of SME employees can substantially increase. This can be seen, for instance, in the US, where SMEs operating in the clothing sector – which incidentally accounts for only 13% of overall employment in the sector – in urban settings are likely to operate in sweatshop-type conditions. According to Sweatshop Watch (SW) – a Californian-based coalition of non-governmental organisations (NGOs) lobbying against sweatshops in the clothing industry – ‘67% of Los Angeles garment factories don’t pay their workers the minimum wage or overtime’.

Lower pay levels in SMEs are also reported in Brazil and India. In the latter country, within the ‘unorganised sector’, pay invariably falls short of the statutory minimum wage set for the various industries and occupations at local level. China, however, appears to be the exception in this context, as no significant differences emerge between the wages paid by SMEs and those paid by larger companies, even if no detailed data are available.

Due to the fact that information on other working conditions, including working hours and the presence and development of HRM practices, is very limited, detailed analysis or comparison is not possible.

Working hours

As for working hours, there is no clear pattern across countries: in certain countries, such as Japan, working hours seem to be positively linked to size (i.e. they are lower in smaller enterprises), while in other countries they are reportedly higher in smaller companies. This is the case with emerging economies such as Brazil and India (but apparently not in China), as well as in the US (though detailed data are not available), where SMEs enjoy greater flexibility as regards compliance with working time regulations. In Europe, longer working hours in SMEs are reported in Belgium, Germany, Greece, Poland and Slovenia, and in France, where the law on the 35-hour week does not apply to firms with less than 20 employees. In other countries, there is no significant differences between working hours in small and large enterprises. But some accounts suggest that this may be due to differences in recording working time, whereby SMEs may more easily and more often register fewer hours than those actually worked.

Job security

An interesting issue concerning SMEs is job tenure/security. While it was not possible to collect data about this with regard to European countries (with the partial exception of Poland where job tenure seems longer in small firms), in countries outside Europe job tenure seems to be significantly higher in larger enterprises. In the US, workers in businesses with fewer than 25 employees have 50% less job tenure than workers those in firms above this threshold (4.4 years compared to 8.5 years). In Japan, a country traditionally characterised by long tenure (especially in large enterprises), the average length of service in 2004 was 10.3 years in companies in the 10-99 employee size group and 15.1 years in companies with 1,000 employees and more. In Brazil, a country where job tenure is particularly short, the proportion of the workforce which stays in the job longer than one year is lower in SMEs: it stands at roughly 50% compared to 67% in large enterprises and 58% on average. Despite the many restrictions faced by SMEs, evidence shows that the labour market for SMEs is relatively more flexible than the average. This may be linked to a number of reasons: often job protection legislation does not apply or is less strict under a certain company size; as career development may be difficult in SMEs, workers are more likely to change and possibly shift to larger companies, therefore work in a small enterprise can provide some skill and experience-building.

HRM practices

HRM practices, as expected, seem to be less present and developed in SMEs. However, for the countries where data is available, it is possible to say that SMEs are not completely devoid of any initiatives in this field; rather, there is a minority of SMEs where innovative HRM practices are indeed present. For instance, a Belgian survey shows that 13% of the firms which can be labelled ‘HRM-rich’ are small firms, whereas in Germany, a research on ‘high-performance work systems’ found that they were present in 11% of firms with up to 19 employees and a third of those with 20-199 employees, which compare with more than half in those with at least 500 employees (Edwards and Ram, 2006). Similar research data exist for Italy, where some 35% of small enterprises (up to 100 employees) present fairly developed HRM systems and 22% combine innovative HRM practice with well-established industrial relations (Bordogna L. and Pedersini R., 2001). European SMEs are also characterised by significant initiatives in training and competence development initiatives. Two Europe-wide surveys found that overall 30% of SMEs carried out continual vocational training in 1999, while 50% of SMEs claimed that competence development was a key part of their business strategies in 2003.

A similar picture emerges in Japan, where, despite the fact that advanced HRM practices is clearly linked to size, a substantial minority of SMEs (between 30 and 99 employees) introduced performance-related pay (47%), performance evaluation systems (26%), personnel appraisal systems (39%), as well as a number of labour-management communication tools, as well as offering training (66%, both off-the-job and systematic on-the-job training). For the US, no precise data are available, but the lower incidence of HRM practices in SMEs is confirmed.

Industrial relations

Industrial relations are remarkably weak in SMEs enterprises. Trade unions often find it difficult to organise in small workplaces, an effort which, apart from the problems which unions may encounter on the field, would require considerable efforts and resources. Both trade union membership and company-level collective bargaining are particularly low in SMEs. Union density in SMEs is often well below the national average: it is estimated at 2% in France, 15% in the Netherlands 13% in Ireland, and less than 2% in Japan. In other countries, such as the US, data on union density in SMEs are not available.

The same low incidence can be found when worker representation bodies are taken into consideration. For instance, in Germany, even when the law allows for the establishment of a works council in firms with at least five employees, data from a 2003 survey show that councils existed in only 7% of establishments with 5–50 employees. A higher presence of representation bodies in SMEs can be found in Spain, where, in 2003, 34% of firms with between 6 and 49 employees had a works council. In the Netherlands, as many as 80% of enterprises with 10–50 employees have a works council. In Japan, a similar influence of size can be found in the diffusion of labour–management consultation bodies (whose forms and entitlements vary in each company): they are present in only around 10% of firms with 30–49 employees compared to almost 50% of those with 300–999 employees and above 60% of those over this threshold.

It is interesting to note that, while the trade union’s side appears to be weak everywhere and collective joint regulation of labour is scarce, the situation of employers’ associations is more diversified and, in certain countries, presents strong and well-established organisations. This can be linked to a number of reasons and especially to the prevailing level of collective bargaining (strong employer associations will be present where multiemployer bargaining is developed), as well as to political and institutional aspects, such as the presence of social dialogue and the commitment to involve SMEs organisations in the definition of economic and social policies.

In these cases, even if industrial relations may be generally weak at company level, the systems of representation and collective bargaining can be significantly developed at national (and possibly territorial) and sectoral levels. This situation exists in countries such as Austria, France, Finland, Italy, the Netherlands, and, Spain where both employer representation and collective bargaining coverage of SMEs are a significant part of national industrial relations systems. This often happens through the inclusion of SMEs in peak or sectoral employer associations and the extension of the respective multiemployer agreements. However, specific representation is also present in some countries, where the employer associations of SMEs engage in social dialogue (as in France and Spain) and also in collective bargaining (as in Italy). A more general aspect, which can be found in virtually all countries covered by this report, is the presence of trade (or business) organisations of SMEs which represent the interests of SMEs vis-à-vis the public authorities and provide a wide array of services to associates.

Due to the weakness of company-level industrial relations, SMEs in countries with company-level bargaining should have significantly lower collective bargaining coverage rates compared to larger companies. Even if specific data are quite limited, it is possible to confirm this tendency for Japan, where more than 80% of SMEs with a trade union were covered by collective bargaining. This means that coverage in terms of employees should roughly correspond to the unionisation rate, which is below 2% in firms with less than 100 employees. The same can be said for the UK, where some sort of collective bargaining takes place only in 4% of SMEs. In the US, small workplaces are rarely organised, so that both union density and collective bargaining coverage is thought to be quite low.

In Brazil, the representation of SMEs is relatively well developed and it is mainly concentrated in local business associations, which have been very active in promoting the interests of SMEs in recent years, especially through supporting the introduction of specific tax and social contribution reductions. Moreover, in principle, collective bargaining coverage of SMEs should be complete, as industry-wide agreements are automatically extended to all establishments in the sector. It could be worth saying that sectoral accords can set different conditions and notably wage increases for distinct firm sizes and often pay rises are lower for SMEs than for large enterprises.

In China, there is no specific representation of SMEs and collective bargaining is virtually non-existent in SMEs. Besides, the elusion of labour laws in SMEs, including on the setting up of trade unions and signing labour contracts, is reportedly higher than in large companies. In India, unionisation and collective bargaining is very limited in the unorganised sector. There are some instances of local union or union-like bodies and forms of collective representation of worker grievances in some clusters of SMEs, but these seemingly represent rare exceptions.


TAW and SMEs are both complex phenomena. The various components of the demand and offer of TAW clearly show how its meaning for user companies and temporary agency workers can be quite different. Similarly, the fragmentation of SMEs, including in the field of employment relations, highlights the multiple roles they can play for the economic systems and their employees. TAW and SMEs are obviously part of distinct aspects of the economic and social environment. While the former directly refers to a specific portion of the labour market, which often involves some of the most critical segments of the labour force in terms of job security and career prospects, the latter represents a key component of the economic structure, which employs generally well above 50% of total employment.

Despite all the differences, in recent years TAW and SMEs have both drawn the attention of experts and policymakers. From rather marginal aspects in the debates and policies about the labour market and the economic systems, they have been brought to the forefront. TAW has no longer been considered only as a frictional short-term flexibility to the almost exclusive benefit of firms, which should have been constrained, in order to limit its possible opportunistic utilisation to elude the rules and protections of permanent employment. Rather, the potential contribution of TAW to labour market entry and the advantages of investing in the employability and career prospects of temporary agency workers have been increasingly recognised. Even more importantly, these advantages have been often acknowledged by TWAs too, as they contribute to strengthen the position of TWAs vis-à-vi s their customers and competitors. Therefore, TAW has started to be taken into consideration in active labour policies, most notably in the EU, but also in other countries such as Japan. Of course, this is a difficult process, which needs to overcome many of the assumptions about TAW by investing in its specific features which can make it attractive for employees.

Similarly, the role of SMEs in economic growth, competitiveness and employment creation has been increasingly stressed. A number of measures have been introduced in recent years in Europe and in other advanced economies to support the viability of SMEs and provide them with assistance in many business domains where they are traditionally weak, including HRM. Some of these initiatives have focused on reducing the burden of administrative requirements and ‘red tape’, while others have tried to make valuable resources and infrastructures available to SMEs in fields such as research, marketing, and finance. Also in emerging countries, policies in favour of SMEs have lately become one of the main instruments to foster economic growth. In terms of the report’s main focus, i.e. the state and development of industrial and employment relations, such efforts to support SMEs are likely to lead to a higher formalisation of HRM and increase the area, which is now limited as this report shows, where innovative practices and possibly developed industrial relations are present.

Of course, these are only general tendencies and the differences across countries should not be underestimated. The meaning of complex phenomena such as both TAW and SMEs depend on the institutional, social and economic contexts in which they are embedded. As we have seen, TAW practically does not exist in China and it is not even an issue for employers, trade unions and policymakers. In India, TAW mainly concerns a very specific part of the economic system (large companies in the formal sector) and therefore remains basically a quite marginal aspect of the labour market. In Brazil, a similar diffusion of informal labour contracts, as well the possibility to dismiss workers ‘with no cause’, contribute to a very limited utilisation of both TAW and fixed-term contracts. Clearly this shows that the demand for TAW by companies is linked to the need for the formal type of adaptability allowed by this form of employment and by the alternatives which are available to companies.

So far, TAW appears to be mainly an issue in advanced economies, where it represents both a source of adaptability for firms and a tool to improve the efficiency and effectiveness of labour markets. Still, the use of TAW can be quite different even in institutional settings which are usually considered as similar: the examples of the US and the UK are interesting in this respect. In the US, the diffusion of TAW is relatively limited, something which is often explained in terms of the considerable degree of labour market flexibility which would make the adaptability allowed by TAW as marginal for companies, while in the UK, which similarly presents a high labour market flexibility, the use of TAW is the highest across the countries considered here and, according to some estimates, involves as much as 5% of total employment. Apparently, quite distinct company attitudes and practices can significantly contribute to the development of TAW, in addition to institutional factors. And this should be taken into account, especially when TAW becomes an instrument to achieve occupational policy goals.

Similarly, SMEs acquire different social and economic relevance across countries. As underlined above, a relevant issue relates to the formal/informal divide, and this, again, points to an important distinction between advanced and emerging market economies: in the latter, the informal economy represents a crucial component of the economic system and a key source of the employment opportunities for workers, with significant consequences in terms of labour protection and regulation and industrial relations. In these countries, support for SMEs is often – even if not necessarily – connected to the formalisation of economic activities and usually part of the efforts to improve the social viability of business initiatives. In advanced economies, while the question of informal activities and labour protections can be similarly important, the focus is more on providing infrastructures, services and incentives to enhance the economic viability of companies and their capacity to produce innovations, including in HRM practices, and cope with increasing international competition.

Despite the increasing attention devoted to SMEs in recent years in all the countries under review here and the introduction of specific policies, a continuing lack of information on this crucial component of economic systems should be acknowledged. Probably, the design of specific policy measures would significantly benefit research and monitoring efforts, in order to fine-tune and adjust existing initiatives. More specifically, in the cases of both TAW and SMEs, the contribution of social dialogue and the social partners to the definition of appropriate policies can be crucial. The challenge is to overcome traditional attitudes and views and make both TAW and SMEs areas where the objective of creating ‘more and better jobs’ can be achieved more effectively.

Roberto Pedersini, University of Milan


Arrowsmith, J., Temporary agency work in an enlarged European Union, European Foundation for the Improvement of Living and Working Conditions, Luxembourg, Office for Official Publications of the European Communities, 2006.

Basque Institute of Research and Studies (Ikea), Employment relations in micro and small enterprises in the EU - literature review, European Foundation for the Improvement of Living and Working Conditions, Luxembourg, Office for Official Publications of the European Communities, 2002.

Bordogna, L. and Pedersini, R., ‘Relazioni industriali e gestione delle risorse umane nelle piccole imprese’, in Giornale di diritto del lavoro e di relazioni industriali, Vol. 23, No. 90, 2001, pp. 209–232.

Edwards, P. and Ram, M., Employment relations in SMEs, EIRO comparative study, European Foundation for the Improvement of Living and Working Conditions, Dublin, 2006.

Goudswaard, A. and de Nanteuil, M., Flexibility and working conditions: A qualitative and comparative study in seven EU Member States, European Foundation for the Improvement of Living and Working Conditions, Luxembourg, Office for Official Publications of the European Communities, 2000.

Micheli, G.A., ‘Svantaggi e benefici del lavoro atipico nel confronto tra domanda e offerta’, in Stato e mercato, No. 78, 2006, pp. 437–472.

Van Jaarsveld, D.D., ‘Collective representation among high-tech workers at Microsoft and beyond: Lessons from WashTech/CWA’, in Industrial Relations, No. 2, April 2004, pp. 364–385.


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

Add new comment

Click to share this page to Facebook securely

Click to share this page to Twitter securely

Click to share this page to Google+ securely

Click to share this page to LinkedIn securely