30   What Makes a Good job?
Georgellis, Y„ J. Sessions and N. Tsitsianis (2003) 'What Makes a Successful Entrepreneur?
An Empirical Analysis of Self-Employment Duration', mimeo (Brunei University). Guest, D. and N. Conway (1999) 'Peering into the Black Hole: The Downside of the
New Employment Relations in the UK', British journal of Industrial Relations, 37,
367-89.
Heywood, J„ W. Siebert and X. Wei (2002) 'Worker Sorting and Job Satisfaction: The
Case of Union and Government Jobs', Industrial & Labor Relations Review, 55, 595-609. Jackson, A. and P. Kumar (1998) 'Measuring and Monitoring the Quality of Jobs and
the Work Environment in Canada', mimeo (Montreal: Queen's University). Leontaridi, R. and P. Sloane (2003) 'Low Pay, Higher Pay, Earnings Mobility and Job
Satisfaction', mimeo (Swansea: University of Swansea). Lévy-Garboua, L,, and C. Montmarquette (2001) 'Satisfaction Judgements and Utility
Analysis', mimeo (Paris: Universitě Paris 1). Locke, E. A. (1976) 'The Nature and Causes of Job Satisfaction', in M. D. Dunnette
(ed.), Handbook of Industrial and Organizational Psycholog}' (Chicago, 111.: Rand-McNally). Lowe, G. and G. Schellenberg (2001) 'What's a Good Job? The Importance of
Employment Relationships', CPRN Study No. W|05. McEvoy, G. M. and W. F. Cascio (1985) 'Strategies for Reducing Employee Turnover:
A Meta-Analysis', journal of Applied Psycholog}', 70, 342-53. McNabb, R. and P. Ryan (1990) 'Segmented Labour Markets', in D. Sapsford and
Z. Tzannotos (eds), Current Issues in Labour Economics (London: Macmillan). OECD (1995) 'Supplementary measures of labour market slack: an analysis of discouraged
and involuntary part-time workers' OECD Employment Outlook (Paris: OECD). OECD (1996) 'Recent Labour Market Developments and Prospects', in OECD Employment
Outlook (Paris: OECD). OECD (1997) Ts Job Insecurity on the Rise in OECD Countries?', in OECD Employment
Outlook (Paris: OECD). OECD (1998) 'Working Hours: Latest Trends and Policy Initiatives', in OECD Employment
Outlook (Paris: OECD). OECD (2003) 'More and Better Jabs? Aggregate Performance During the Past Decade1,
in OECD Employment Outlook (Paris: OECD). Patterson, M., M. West, It. Lnwthorn and S. Nickeil (1997) 'Impact of People Management
Strategies on Business Performance', Issues in People Management No. 22 (London:
Institute of Personnel and Development). Rosenthal, N. (1989) 'More than wages at issue in job quality debate', Monthly Labor
Review, 112, 4-8.
Sousa-Poza, A. and A. A. Sousa-Poza (2000) 'Taking Another Look at the Gender/
Job-Satisfaction Paradox', Kyklos, 53, 135-52. Stutzer, A. and B. Frey (2003) 'Stress That Doesn't Pay Off: The Commuting Paradox',
1EER Working Paper No. 151 (Zurich: University of Zurich). Tsuru, T. and J. Rebitzer (1995) 'The Limits of Enterprise Unionism: Prospects for
Continuing Union Decline in Japan', British journal of Industrial Relatiuns, 33, 459-92. Van Praag, B. M. S., and B. Ľ. Baarsma (2004) 'Using happiness surveys to value
intangibles: the case of airport noise', Economic journal, Forthcoming. Warr, P. B. (1999) 'Well-Being and the Workplace', in D. Kahneman, E. Diener and
N. Schwarz (eds), Well-Being: The Foundations uf Hedonlc Psychology (New York: Russell
Sage Foundation).
2
Job Quality in European Labour, Markets
Frank Skbem-Thomas*
Introduction
Between 1997 and 2002 more than 12 million new jobs were created in the European Union (EU) and labour market participation increased by more than eight million.1 According to recent analyses {European Commission, 2001a, 2002, 2003a) the majority of the new jobs were highly skilled ones in high-tech and/or knowledge-intensive sectors, offering decent pay, job security, training and career development. At the same time, however, in some sectors employment growth was stronger for temporary or low-paid jobs than for permanent, highly paid ones. The employment share of people in temporary jobs reached almost 14 per cent in 2000, and that of people in low-paid jobs around 20 per cent.
In fact up to a quarter of EU workers hold jobs of comparatively low quality due to low pay and productivity, job insecurity and lack of training, and thus are at disproportionate risk of unemployment and social exclusion. Together with the continuing destruction of low-skill, low-productivity jobs, this not only highlights the difficulty of integrating individuals with few skills into the labour market, but also raises questions about the long-term sustainability of recent employment trends and their potential downside in terms of job quality.
Moreover it is questionable whether there are synergies between quality and quantity of employment - that is, whether improvements in the quality of work favour employment creation and productivity improvements, or whether there are trade-offs between the two. And while increased flexibility in the labour market may favour job creation and the adjustment of the economy to cyclical fluctuations, its impact on other policy objectives - such as improved productivity and social cohesion - remains unclear.
*The views expressed in this chapter are those of the author and do not necessarily correspond to those of the European Commission.
32  fob Quality in European Labour Markets
Frank Siebern-Tltamas 33
This chapter lays out the policy background for work quality and gives a summary overview of job satisfaction, quality of work and quality dynamics in the EU, based on the European Commission's annual Employment in Europe reports. It also discusses the relationship between quality of work and the two other overarching objectives of the EU employment strategy: full employment and social inclusion.
Quality of work: political background and monitoring
The concept of job quality is not new.2 Job satisfaction was introduced in the sociology and economics literature as early as the 1970s, and it was also taken up in the policy debate, notably in Canada and the United States.3 In the latter there has been a long-standing discussion among academics and policy makers on the measurement of job quality and its link with overall employment performance.'1
Various institutions have developed monthly or quarterly employment quality indices to track changes in job quality over time. These indices are calculated on the basis, among other things, of job characteristics such as wages, benefits and job stability. They are used in studies of the type of employment created in the United States, where many high-quality jobs have been replaced in recent years by less well paid and less stable ones (Tal, 2004).
The International Labour Organisation (ILO) brought the concept of job quality to the forefront in its 'decent work' agenda (ILO, 1999, 2003). The OECD on the other hand, despite repeated analyses of the concept (for example Clark, 1998; OECD, 2001, 2003a, 2003b), has generally restricted its attention to improvements in labour force skills and competences.
From a European policy perspective, the concept of job quality rose to prominence at the Lisbon European Council in March 2000, when the EU's strategic goal for the next decade was defined: to become the most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion. This moved EU employment policy beyond the social protection, health and safety and equality agenda that had influenced it since the Social Chapter came into being in 1961.
The new focus on quality of work - better jobs - was reaffirmed at the European Council meetings in Nice in 2001 (which included the promotion of quality as a main theme for actions and initiatives in the period 2000-5 in the European social agenda), Stockholm in 2001 (which also considered the introduction of quality as a general objective in the employment guidelines), Barcelona in 2002 (which recognized that quality would facilitate higher employment levels and that the objective of creating better jobs would complement and reinforce that of creating more jobs) and Brussels in 2003 (which emphasized that the objectives of raising employment rates
and improving the productivity and quality of work were interrelated and mutually supportive). Improving the quality of work - one of the three main objectives of the European employment strategy, together with full employment and social cohesion - is thus seen as important not just for the well-being of workers but also to promote social inclusion and drive up employment levels.
In the absence of a single, widely accepted definition of job quality5 the European Commission identified 10 dimensions of job quality in a communication in 2001 (European Commission, 2001b; for an overview see Table 2.1 below), taking account of the fact that the concept of quality of work is
Table 2.1 Dimensions of quality in work and quality indicators Dimension Description
Intrinsic job quality
Skills, lifelong learning and career development
Gender equality
Health and safety at work
Flexibility and security
inclusion and access to the labour market
Jobs ought to be intrinsicaliy satisfying, compatible with a person's skills, abilities and expectations, and provide appropriate levels of income.
People ought to be able to develop their potential abilities to the full through access to the knowledge society and appropriate support for life-long learning in general and training in new technologies in particular.
Labour markets should offer equal opportunities for men and women in terms of lifetime careers, including equal access to employment and supervisory functions, and equal pay for work o£ equivalent value to society.
Working conditions must be safe, healthy and supportive (in both physical and psychological terms) of sustainable participation and employment. Exposure to risks and stress at the working place, accidents at work, and occupational diseases must be reduced.
An appropriate balance between flexibility and security of employment and work relationships is required to encourage a positive attitude towards change in the workplace and the labour market. This will involve appropriate support for those who lose their jobs or are seeking alternative ones, as \vell as encouraging the full use of abilities and flexible career choices through appropriate support for occupational and geographical mobility.
Labour markets should be open to all, including those entering the labour market for the first time or after a period of unemployment or inactivity. No one should be permanently excluded from the labour market due to low educational attainment, inadequate skills, previous unemployment or inactivity.
34 lob duality in European Labour Markete
Frank Siebern-Tliumas 35
Table 2.1 (Continued)
Dimension
Description
Work organization and work-life balance
Social dialogue and worker involvement
Diversity and non-discrimination
Overal! work performance
Working arrangements, especially in respect of working hours, and support services should allow an appropriate balance between working life and life outside work. All workers should be informed about and involved in the development of their jobs and companies through appropriate industrial relations activities and guaranteed rights and representation.
All workers should be treated equally and without discrimination in terms of sex, age, disability, ethnic origin, religion or sexual orientation. High levels of labour productivity and high living standards across all regions of the EU should be targeted.
Source: European Commission (2001b).
multifaceted and encompassess a range of characteristics, such as job, worker and firm characteristics; the subjective evaluation of these characteristics by the worker on the basis of his or her characteristics, experience and expectations; the quality of the job-worker match; the working environment; and the direction and priorities of employment and social policies. For each of the 10 dimensions, one or more indicators were subsequently proposed - and adopted at the Laeken summit in December 2001 - as a means of assessing the quality of work in Europe and of monitoring its evolution over time, most notably in the framework of the European employment strategy (European Council, 2001).6
On the basis of these indicators the European Commission reviewed the efforts being made to improve the quality of work in the member states and assessed the progress made to date. It concluded that while 'recent trends and performances around these ten dimensions were encouraging in some respects, overall there was scope for considerable improvement under each of the ten dimensions of quality.' (European Commission, 2003b). Furthermore, while a few of the member states 'perform well under most indicators of quality (Denmark, the Netherlands, Sweden, Austria), others display consistently much less favourable performances (Italy, Greece, Spain, Portugal). The picture is more mixed for the remaining member states' (ibid.)
With regard to employment policies, the Commission stressed that there had been
considerable progress over recent years in several policies supporting quality in work. In particular activation and prevention policies have been reinforced since 1997 in response to the employment guidelines,
aiming at smoother transitions from unemployment and inactivity to employment. Steps have also been taken in most Member states to reduce unemployment and poverty traps through changes in the tax/benefit systems, in particular for the young, older and low-skilled workers, who are at a higher risk of unemployment or inactivity, (ibid.)
These improvements notwithstanding, the report concluded that
more determined policy action is needed especially with respect to encouraging investment of firms in training and promoting active ageing through better adapted working conditions and greater financial incentives to remain in work. At the same time, wider access to care services for children and other dependants must be provided while efforts should be made in substantially reducing accidents at work and occupational diseases. A strong involvement and commitment of the social partners is a necessary condition and key factor of success for improving quality in work. (Ibid.)
Analysing job satisfaction and quality in work in Europe
Given the above definition of job quality and the absence of a single composite indicator of job quality, the empirical analysis reported here had to be based on objective data on job and worker characteristics and subjective evaluations of the job-worker match. The first step was to analyze individuals' self-reported satisfaction with their employment status (employed, unemployed, inactive) and, for the employed, with their job in general, its specific characteristics (such as earnings, job security, working time, working hours, work content, work control, working conditions and work-related health) and its main determinants.
In the second step an analytical concept was derived to identify 'good' and 'bad' jobs. Jobs were classified into four types - good jobs, jobs of reasonable quality, low-paid jobs, and dead-end jobs - according to objective characteristics such as job security, work content, training possibilities, career prospects, productivity and pay (these characteristics have also been found to have the strongest impact on individuals' self-reported job satisfaction).
Next, transitions between jobs of different quality were analyzed to assess labour market integration prospects and changes in job quality on the one hand, and vulnerability to job loss and social exclusion on the other. Finally, the part played by quality of work in overall employment performance was analyzed by means of a dynamic simulation.
The following subsections review the main results of these analyses, based on data for the years 1995-2000. Details of the main data sources are presented in Appendix 2.1. Due to the lack of recent comparable data on the new EU member states the analysis is restricted to the EU-15.
36 Job Quality in European Labour Markets Self-reported satisfaction levels
European Community Household Panel (ECHP) data enable us to analyze two types of self-reported satisfaction. First, the respondents in the 2000 survey were asked to state their satisfaction with their employment status (employed, unemployed or inactive). This is of particular interest when analyzing the voluntary/involuntary nature of labour markets and the determinants of labour market participation. Second, all those in employment were asked about their satisfaction with their job in general and with its specific characteristics: earnings, job security, working time, working hours, work content, work control, working conditions and work-related health.
Satisfaction with employment status
Almost three quarters of the respondents were rather satisfied with their employment status, the rest declared themselves to be rather dissatisfied.7 Twelve per cent were fully satisfied, while 5 per cent were not satisfied at all. Overall satisfaction was highest in Denmark, Austria and the Netherlands, where two thirds or more claimed to be satisfied and less than 5 per cent expressed dissatisfaction. Dissatisfaction was higher in Italy (22 per cent), Greece (18 per cent), Portugal (14 per cent), Spain (13 per cent) and the UK (12 per cent).
Not surprisingly, in all member states employment and inactivity were rated much more highly than unemployment. While 80 per cent of the employed and 70 per cent of the inactive respondents were rather satisfied with their status, 70 per cent of the unemployed were rather dissatisfied, and almost half of the latter were not satisfied at all. With the exception of Denmark and the Netherlands, in all the member states at least 50 per cent of the unemployed (and many more among young unemployed people) were highly discontent, most notably in Italy, Greece and Portugal, where more than 80 per cent of the unemployed expressed strong dissatisfaction.
Self-reported satisfaction among inactive people varied markedly according to personal characteristics. First, while younger, older and highly skilled individuals tended to express a high degree of satisfaction with not being in the labour market (usually due to being in education in the case of the young and retirement in the case of older people), women and low-skilled individuals were strongly dissatisfied with being inactive. Second, while in many of the member states the satisfaction levels of the inactive were comparable to, if not higher than, those of the employed, in countries witlT low female participation in the labour force (Ireland, Italy, Greece, Spain and Portugal), on average the satisfaction levels of the inactive were well below those of the employed. Both these findings were in line with the relatively high share of unemployment or inactivity among women, the young and the low-skilled, who generally reported a strong willingness to take up work in the near future.
Frank Sieben i-Thomas 3 7
Changes in the level of satisfaction with employment status were driven predominantly by transitions into or out of unemployment. Transitions into unemployment were linked to decreasing satisfaction levels, irrespective of the country in question, while transitions out of unemployment into employment or inactivity generally increased the reported levels of satisfaction.
fob satisfaction
Almost 80 per cent of the employed respondents declared themselves to be rather satisfied with their current job (12 per cent were fully satisfied). Most of the remainder were rather dissatisfied, and 3 per cent were not satisfied at all. When asked to evaluate specific job characteristics the respondents generally expressed greater than average satisfaction with work content, working times and working conditions, and lower than average satisfaction with job security, working hours and earnings.
Although cross-country differences generally have to be interpreted with caution, there were significant differences in self-reported job satisfaction across the EU countries. Job satisfaction was highest in Denmark, Austria, the Netherlands and Belgium, where more than 60 per cent of the employed declared they were satisfied, but less than 40 per cent in Italy, Greece and Portugal. Dissatisfaction rates of 10 per cent or more were found in Greece, Italy and the UK (Figure 2,1).
100
75
50
25
0
Den. Aust. Neth. Belg. Ire.   Fin.   UK   Fr.   EU   Sp.    It.    Qr. Port.
CD Satisfied     Intermediate □ Dissatisfied
Note: No data are available for Germany, Luxembourg and Sweden. Source: European Community Household Panel, wave 7 (20TJ0).
Figure 2.1  Self-reported job satisfaction, by member state, 2000 (per cent)
38  lob Quality in European Labour Markets
Frank Siebem-Tltamas 39
Job satisfaction varied considerably according to personal characteristics and to job characteristics such as working time, contract type,8 size of employer, sector and occupation. Despite differences among the member states in the determinants of job satisfaction, these were some common patterns, job satisfaction tended to be higher among those in voluntary part-time jobs, on permanent contracts, in supervisory positions and with access to training at the workplace (Figure 2.2). In particular, voluntary part-time workers generally expressed the highest satisfaction with their working hours, working times, work content and working conditions. On average those in part-time work because of childcare or other family responsibilities or because they were engaged in training or education reported higher job satisfaction levels than the full-time employed. Job satisfaction also tended to be significantly greater among those in highly-skilled, non-manual occupations in the service sectoT in general, and in education, health and social services and the public sector in particular.
In contrast temporary contract workers, workers without access to training and involuntary part-time workers expressed strong dissatisfaction. Their discontent was not restricted to working hours, job security or earnings but also included work content, which suggests that they would have been happier with an entirely different job. Some dissatisfaction was also reported by low-skilled manual and unskilled workers and by workers in agriculture, construction, manufacturing, hotels and restaurants (Figure 2.3). Dissatisfaction was particularly marked in the case of work content and working conditions. However there were large differences within sectors, notably in
25
50
75
100
Wark place training Supervisory position High-quality Job Intermediate Job status Voluntary part-lime Part-time (family responsibilities) Part-time Permanent contract Average Low-paid |cb FulMime Part-time {In education) Non-supervisory |ob Temporary contract No work place training Involuntary part-time Dead-end Job
□ Satisfied 11 Intermedials □Dissatisfied
Nole: There are no data for Germany, Luxembourg and Sweden. Source: European Community Household Panel, wave 7 (2000).
Figure 2.2 Self-reported job satisfaction, selected job characteristics, 2000 (per cent)
25
50
Education Health and social wort; Public administration ElBctricity, gas and water supply Real estate and business activities Services total Financial intermediation Mining and quarrying Wholesale and retail trade Transport and communication Construction Manufacturing Hotels and restaurants Agriculture
75
100
I □ Satisfied S3 Intermediate □ Dissatisfied]
Nole: There are no data for Germany, Luxembourg and Sweden. Source: European Community Household Panel, wave 7 (2000).
Figure 2.3  Self-reported job satisfaction, by sector, 2000 (per cent)
manufacturing, where satisfaction was much higher among respondents who worked in factories producing metal products or chemicals than among those in the textile industry, where satisfaction was even lower than in agriculture. With regard to individual jobs, the greatest dissatisfaction was expressed by agricultural workers and people with elementary occupations {Figure 2.4).
0
25
50
Professionals Legislators Technicians Clerks Service workers Craft workers Plant operators Elementary occupation Agricultural workers
75
100
[□Satisfied HIntermediate □ DissatisfiedI
Note: There are no data for Germany, Luxembourg and Sweden. Source: European Community Household Panel, wave 7 (2000).
Figure 2.4 Self-reported job satisfaction by occupation, 2000 (per cent)
40 Job Quality in European Labour Markets
Frank Siebeni-Thoinas 41
The above findings are confirmed by econometric analyses of the main factors associated with higher job satisfaction (see Appendix 2.1). In all member states, in 2000 job satisfaction was positively correlated with wages, job status and job-related skills acquired through training, and negatively correlated with temporary contract work, job-worker mismatch and over-qualification. Job satisfaction was significantly higher in the service sector in general and the public sector in particular, and among those employed in highly skilled non-manual occupations such as legislators, managers, professionals and technicians. In contrast men in the construction sector and both men and women in elementary occupations reported significantly lower levels of job satisfaction.
On the basis of the above analyses, two interesting observations can be made. First, when controlling for job-related, sectoral and occupational characteristics, low-skilled workers do not necessarily have a lower degree of job satisfaction. On the contrary, a small but significant negative effect is only found for the highly skilled, pointing to a potential mismatch between job characteristics and the jobholder's expectations. Second, even after controlling for the above characteristics there is evidence that the gender concentration in a sector - that is, the share of women employed in that sector - may have a negative effect on job satisfaction, often in addition to the strong wage penalties in such sectors.
Significant changes in job satisfaction occur after job changes and pay increases. In general a move out of industry or agriculture into the service sector is related to greater satisfaction, possibly because of the higher quality of jobs in the service sector. A similar increase in job satisfaction results from promotion to a higher status, and from moving from an atypical form of work to a more standard one, such as from a temporary to a permanent contract or from involuntary part-time work to full-time employment
Identifying 'good' and 'bad' jobs in Europe
Based on the finding that contractual insecurity, low pay/low productivity, lack of responsibility and the absence of career development opportunities are among the main causes of job dissatisfaction in the EU, the European Commission (2001a) classified jobs into four categories: j
• Dead-end jobs: non-supervisory jobs on either fixed-term or short-term contracts or without a formal contract; there is no employer-provided training.
• Low pay/productivity jobs: jobs paying hourly wages below 75 per cent of the country median but offer job security and/or access to employer-provided training. Hourly wages below 75 per cent of the country median indicate that these jobs are ones of relative low productivity,9
• Jobs of reasonable quality: jobs with decent pay and productivity above 75 per cent of the country median, plus either job security or access to employer-provided training.
• Good jobs: jobs with all the characteristics listed in the preceding point.
According to this classification, 73 per cent of all jobs in the EU in 2000 were of good or reasonable quality. Thus 27 per cent of the workforce were in low-quality jobs.10 Of these jobs, roughly a third were without job security or employer-provided training (and therefore were precarious and offered no career prospects), and a half of this third were of low pay/ productivity. The other two thirds, despite their low pay/productivity, at least offered job security and/or access to training, although a large proportion of them could be characterized as dead-end. The highest share of low-quality jobs was accounted for by temporary contract workers, and especially temporary workers in part-time jobs.
Job quality, as defined above, varies considerably among the EU member states (Figure 2,5). In 2000 the proportion of national jobs of comparatively good quality ranged from 81 per cent in Denmark to 61 per cent in Spain, and that of dead-end jobs from 3 per cent in Denmark to 27 per cent in Spain. Countries with an above average proportion of good and reasonable jobs were Denmark, Austria, Belgium, Finland, the Netherlands, Italy and France. In the first five of these countries at least 50 per cent of the jobs were
Den. AusL   It.    Fin. Belg. Neth. Fr.   EU  Ger.  Ire.   UK Port.  Gr. Sp.
■ Good El Reasonable ID Low pay/productivity □ Dead-end
Note: For a defenition of the job quality categories see European Commission (2001a). See also Note 10,
Source: European Community Household Panel, wavB 7 (2000). pJSure 2.5 Job quality in the EU, by member state, 2000 (per cent)
42  fob Qua lit)' in European Labour Markets
Frank Siebem-Thomas 43
classified as good. Dead-end jobs, on the other hand, were above the EU average (8.5 per cent) in Spain, Greece, Portugal, Germany, Italy and Ireland the shares of dead-end jobs cannot be estimated for France and the UK.n
Job quality also differs significantly according to the characteristics of the individual job holder, particularly in respect of gender, age and educational background. Young and low-skilled workers are likely to be in jobs with low pay and productivity and that lack job security, training opportunities and career prospects. In fact up to 18 per cent of young employed people and around 10 per cent of the low skilled are in dead-end jobs. Women also are overrepresented in low-paid jobs.
Another variation in job quality is related to working time. While a quarter of full-time jobs and a third of voluntary part-time jobs are of relatively low quality (low pay/productivity or dead-end - Figures 2.6 and 2.7), over two thirds of involuntary part-time workers are in such jobs (Figure 2.8). Given that these workers would prefer a full-time, better job, the lack of suitable employment opportunities is worrying.
Clear differences in job quality also exist across occupational groups and sectors. Half or more of low-skilled or unskilled manual jobs are of rather low quality, with 10 per cent of service workers and more than 20 per cent of agricultural workers and workers in elementary occupations occupying dead-end jobs (Figure 2.9). Almost 40 per cent of service workers and shop assistants also receive low pay. On the other hand a very large proportion of highly skilled non-manual jobs are of good or reasonable quality.
16%
Notes: For a defenitlon of the fob quality categories see European Commission (2001 a). See also Note 10.
Source: European Community Household PanBl, wave 7 (2000). Figure 2.6 Job quality in the EU, full-time jobs, 2000
Reasonable 39%
Low pay / produciivity 20%
Nate: For a defenition of the Job quality categories see European Commission (2001 a). See also Note 10.
Source: European Community Household Panel, wave 7 {2000}. Figure 2.7 Job quality in the EU, voluntary part-time jobs, 2000
Reasonable 26%
Low pay/productivity 41%
Note: For a defeniiion of the job quality categories see European Commission (2001 a}. See also Mote 10.
Source: European Community Household Panel, wave 7 (2000). Figure 2.8 Job quality in the EU, involuntary part-time jobs, 2000
At the sectoral level, 28 per cent of jobs in the service sector are good jobs and 27 per cent are of low quality, while in industry 22 per cent are good and 23 per cent are low-quality (Figure 2.10). The higher share of low-quality jobs in the service sector is due to its higher share of low-paid jobs; the share of dead-end jobs is the same in industry and services (8 per cent). In agriculture
44
Frank Siebem-Thomas 45
25
50
75
100
legislators Professionals Technicians Clerks Craft workers Plant operators Service workers Elementary occupation Agricultural workers
□ Jobs of good or reasonable quality 0 Low-paid / low productive jobs
□ Dead-end jobs
Notes: For a defenition of the job quality categories see European Commission (2001a). See also Note 10.
Source: European Community Household Panel, wave 7 (2000). Figtire 2.9 Job quality in the W, by occupation, 2000 (per cent)
0
25
50
75
100
Electricity, gas and water suppiy Financial sen/ices Mining and quarrying Public administration Education Manufacturing Transport and communication Real estate and business activities Health and social work Services total Construction Wholesale and retail trade Agriculture Hotels and restaurants
[3 Jobs of good or reasonable quality d Low-paid / low productive jobs □ Dead-end jobs
Notes: For a deienition of the job quality categories see European Commission (2001a). See also Note 10.
Source: European Community Household Panel, wave 7 (2000). Figure 2,10 Job quality in the EU, by sector, 2000 (per cent)
almost 60 per cent of jobs are of relatively low quality, almost a third are dead-end and a quarter are classified as low pay characteristics / low productivity.
With regard to intra-industry differences, good jobs account for almost 90 per cent of jobs in the energy sector and more than 80 per cent in the manufacturing of metal products and chemicals, while 35 per cent of Jobs in the construction sector are low-quality on dead-end, and 30 per cent or more of people employed in the manufacturing of food products and textiles are in low-paid, low-quality jobs. In the service sector, 80 per cent or more of jobs in education, public administration and financial services are high-quality, but in the retail trade 40 per cent of jobs are low-quality, and in hotels and restaurants 64 per cent are low-quality and more than 16 per cent are dead-end.
Factors contributing to the high share of low-quality jobs in the sectors listed above, are the poor provision of training, the large number of employees on temporary contracts and the generally low pay, with the average wage in some of these sectors being more than 40 per cent below the national average. In Germany and the UK, for instance, wages in hotels and restaurants are among the lowest in the enlarged EU.12
D'Addio urn/. (2004) have obtained similar results on the determinants and variation of job quality in their cluster analysis study - allotting groups of fobs to different clusters according to their objective quality without imposing the determining characteristics beforehand. According to their analysis, contract type, access to training and occupation are distinctive features of job types. People employed in elementary occupations, craft workers, machine operators and service workers are most likely to be in low-quality job clusters. The impact on job quality of variables such as working time, firm size and work in the public sector is different for men and women.13
D'Addio etal.'s results also largely match those on self-reported job satisfaction, with the latter clearly varying across job quality categories. Almost 60 per cent of those employed in good jobs in 2000 expressed high a degree of job satisfaction, as opposed to the less than 5 per cent who declared ■ themselves to be dissatisfied. Job satisfaction was highest among young and highly educated people in good jobs, 65 per cent of whom stated that they were satisfied or very satisfied. In the case of low-paid jobs, about 50 per cent of employees were quite satisfied while 10 per cent were very dissatisfied. The least satisfaction was reported by those in dead-end jobs, of whom about 20 per cent were very dissatisfied. This notwithstanding, 30 per cent of all those in comparatively low-quality jobs claimed they were highly satisfied.
; The correlation between objective job quality and subjective job satisfaction is corroborated by a regression analysis of the main determinants of job satisfaction (see Appendix 2.1). Even controlling for pay levels, contract status and job status, people in jobs of relatively high objective quality
46  lob Qiialit)' in European Labour Markets
Frank Siebern-Thomas 47
reported significantly higher levels of job satisfaction. The only exception is related to the size of establishment. While objective job quality and earnings are considerably lower in small firms, there is some evidence that job satisfaction in small and medium-sized firms is higher than in larger organizations. This is probably due to workers' more varied activities and greater responsibility, better and more transparent work organization, better employee participation in information sharing and consultation, and better personal relationships between workers and management, other possible factors are social dialogue and worker involvement, work organization, a good work-home life balance and good health and safety at work, dimensions of job quality that could not be included in the analysis due to lack of data.
Quality of work and labour market dynamics
As stated earlier, access to the labour market and the existence of career development opportunities are important elements of quality in work. They can best be monitored through labour force changes and their determinants, including moves between the main employment statuses (employed, unemployed, inactive), moves to different employment arrangements and job characteristics (employed versus self-employed, full-time versus part-time, temporary contract versus permanent contract, low pay versus high pay, low-quality versus high-quality work and so on), and income or earnings mobility, as measured by moves between different quintiles of income or earnings distribution. Detailed analyses of these can be found in European Commission (2002a); this section will concentrate on labour force changes by job quality status.
Such transitions are generally characterized by relatively high persistence of both inactivity and high-quality jobs of around 85-90 per cent, as well as low outflow rates from high-quality jobs into unemployment. There is generally little change in the status of those employed in jobs of high quality, with only around 5 per cent experiencing a deterioration of job quality over the years. The situation of the unemployed and those in low-quality jobs, on the other hand, is much more changeable. Around 25 per cent of people who have been unemployed for up to a year move into employment, 66 per cent into jobs of low quality and about 33 per cent into high-quality ones. At the same time a significant minority of the unemployed (up to 20 per cent) withdraw from the labour market altogether. Of those who have been unemployed for more than a year, 64 per cent remain unemployed and up to 20 per cent move into employment, 75 per cent of these into jobs of low quality (8,5 per cent dead-end jobs, 6 per cent low-paid jobs).
Over a periods, two-year on average 60 per cent of people employed in a low-quality job remain in that job category, while 33 per cent experience an improvement, either by achieving job security or gaining access to training and career opportunities, or through an increase in pay. Persistence in jobs of low quality is highest among the low-skilled and the young, while
upward movement is most likely for the highly skilled. Persistence in low-paid employment is also relatively high, with more than 50 per cent of those with low pay staying in poorly paid, low-productivity jobs. This applies particularly to women, with highly skilled men yet again more likely to move into high-quality jobs.H
People in dead-end jobs or low-paid, low-productivity ones are much more likely to withdraw from the labour force or to become unemployed than those in jobs of relatively better quality. This applies not only to the low-skilled but also to older workers. While working in a dead-end job implies a much higher risk of unemployment for all groups, women and the young face the greatest risk, of withdrawal from the labour force, regardless of their skill level. Over a five-year period the rate of transition from low-quality into high-quality employment increases, with up to 50 per cent moving upwards. At the same time, however, more than 20 per cent of people in jobs of lower quality leave employment. This is twice as many as those who leave from high-quality jobs.
The probability of being laid off or experiencing a deterioration of job quality is at least twice as high for people with low-quality jobs than for those who move up the job-quality ladder. The risk of moving out of employment is also pronounced for individuals who have left unemployment by taking a job of low quality or temporary work. Almost 33 per cent of them return to unemployment within a year, as opposed to around 20 per cent of those in permanent, low-paid, low productivity jobs and around 10 per cent of those in jobs of high quality. In addition, unemployed people who take a job without training opportunities are twice as likely to return to unemployment as those who take jobs with access to training. Moreover, moving from unemployment into a job without training opportunities increases the probability of complete withdrawal from the labour force by almost a factor of three. Those who move back into the labour market after a period of inactivity face a similar probability. Withdrawal after a year is greatest among people in dead-end or low-paid, low productivity jobs, but virtually non-existent among those in jobs of high quality.
There are notable differences in the patterns of employment change across EU member states. In 1999-2000 the rate of transition from low- to high-quality jobs ranged from 25 per cent or less in Greece, Portugal and the UK to 30 per cent or more in Belgium, Italy, the Netherlands and Ireland (Figure 2.11). Over a period of five years or longer the figure rises to 50 per cent or more for Ireland, Denmark, Finland, Belgium, Austria and the Netherlands but remains below 40 per cent for Spain, Germany, Greece, France and, most notably, the UK (Figure 2.12).
In France, Spain, Germany and Greece, over longer periods the transition fate is negatively affected by the relatively high outflow from low-quality jobs into unemployment - despite sometimes relatively favourable annual upward dynamics - and a low transition from unemployment into employment.
48
49
El Inactivity H Unemployment □ Low-quality jobs S High-quality jobs
NoSes: In the interest of clarity the rates of transition into unemployment or inactivity are presented as bars to the left in the figure. No data are available tor Sweden and Luxembourg. Source: European Community Household Panel, waves 6 and 7 (1999 and 2000).
Figure 2.11 Transition out of low-quality employment into inactivity, unemployment, low-quality jobs or high-quality jobs, 1999-2000 (per cent)
H Inactivity     H Low-quality jobs    □ High-quality jobs
Nofas: In the interest of clarity the rates of transition into unemployment or inactivity are presented as bars to the left in the figure. No data are available for Sweden and Luxembourg. Source: European Community Household Panel, waves 5 and 7 (1999-2000).
Figure 2.13 Transition out of unemployment into inactivity, low-quality jobs or high-quality jobs, 1999-2000 (per cent)
30     20 10
10      20     30     40     50      60     70 80
ä Inactivity ED Unemployment □ Low-quality jobs El High-quality jobs
Notes: In the interest of clarity the rates Df transition into unemployment or Inactivity are presented as bars to the left in the figure. No data are available for Sweden and Luxembourg. Source: European Community Household Panel, waves 2, 6 and 7 (1995,1999 and 2000).
Figure 2.12 Transition out of low-quality employment into inactivity, unemployment, low-quality jobs or high-quality jobs, 1995-2000 (per cent)
I
20
15
. a
E >,
E
dj
■ c
a io
Spain* Greece France	• Finland • Germany ^
Portugal • •UK	EU average .Denmark Ireland •„ .."Belgium . Austria Netherlands
10 15 20 25 30
Transition into high-quality employment
35
40
Notes: No data are available for Sweden and Luxembourg.
Source: European Community Household Panel, waves 2-7 (1999-2000).
Fi$we2,14 Average transition rates out of low-quality employment, 1995-2000 (per cent)
50  job Quality in European Labour Markets
Frank Siebern-Thomas  51
A high rate of transition into inactivity, however, does not seem to affect long-term patterns in the same way, as can be seen in the case of Ireland, Denmark and the Netherlands, where there is evidence of large fluctuations between employment and inactivity. Interestingly, these are also the only countries, together with Austria, in which the majority of those who move from unemployment to employment do so by taking a high-quality job (Figure 2.13). The UK is a case apart in that low-quality jobs in that country are associated with a comparatively low risk of unemployment and the lowest probability of all EU member states of a move to a better job - hence workers are to some extent locked into these jobs (Figure 2.14).
The above findings indicate that in addition to the relatively high persistence of low-quality jobs there is a close link between job quality on the one hand, and unemployment, social exclusion and poverty on the other. This link is also due to both duration dependence of labour market transitions and employment prospects.'5
The link between quality of work and employment performance
The previous sections have shown that for some subgroups in the labour market - notably low-skilled and older workers - low-quality jobs do not necessarily serve as stepping stones to more stable and more productive employment, although they can be an effective means of reintegrating young and highly skilled people into the labour force. In the case of the former group, it seems necessary to combine measures to reintegrate them into the labour force with measures to help them stay - and develop - in stable employment. It has also been shown that there is a significant link between job quality and labour market participation, and that improvements in the quality of jobs are needed to entice inactive people back to the labour market. However some commentators claim that quantity-quality trade-offs might exist and that quality improvements can have negative effects, with such improvements being taken as equivalent to increases in labour costs or obstacles to hiring/firing and/or wage flexibility. It is therefore important to consider how job quality and upward mobility affect productivity and overall employment performance.
A static view
According to the European Commission (2002), there is evidence in the EU of a positive link between job quality and overall employment performance.1" A comparison of the activity rates and employment rates in countries with a relatively low share of workers in low-quality jobs (below 25 per cent) with those with a relatively high share (above 25 per cent) reveals a positive correlation between the share of high-quality jobs and the activity and employment rate. This positive correlation is particularly marked for women and for the medium- and highly skilled. The employment rate for low-skilled
y:,..   individuals is generally low in all member states, but even more so in those .   with a high share of low-quality jobs. There is also evidence that recent !>: J   increases in the activity and employment rates have been accompanied by r      job-quality improvements in many member states.
j-;:       Labour productivity is related to both job quality and job satisfaction. :   Labour productivity is significantly above average in sectors that offer above average job quality, and considerably below average in sectors with a rela-j; ■.:   tively high share of low-quality jobs. This positive quality-productivity link
* can be observed in all countries for which data are available. In terms of national averages, labour productivity in sectors with relatively high-quality
t      jobs ranges from 110 per cent in the UK to more than 130 per cent in Italy and Austria. In sectors with relatively low-quality jobs' productivity levels ■.  range from 85 per cent in the Netherlands to 55 per cent in Austria, While causality is likely to run in both directions - with higher productivity causing higher job quality, and vice versa ~ it is important to note that a similar positive relationship can be found between actual labour productivity and self-reported job satisfaction. In terms of national averages, labour productivity :. in sectors with relatively high self-reported job satisfaction levels ranges from around 110 per cent in the Netherlands, Denmark and France to more j       than 125 per cent in Austria. In sectors with relatively low self-reported job |      satisfaction levels it ranges from 90 per cent in Denmark to less than 70 per cent in Italy and Austria.
!; At the sectoral level, labour productivity is also correlated with the various ; components of job quality defined earlier, particularly in the case of access to training, training incidence and contractual job security. As discussed ! above, in some sendee sectors with low productivity, such as hotels, restaurants j and the retail trade, training provision is generally low and a large proportion If of employees are employed on a temporary basis, thus contributing to the £ relatively high share of around 15 per cent of people employed in dead-end |. jobs and up to 50 per cent in low-paid, low productivity jobs. High-quality K; jobs account for 80 per cent or more of total jobs in financial services, transport I      and communication and the energy sector - three of the most productive
• sectors in the European economy.
Dynamic simulations
Notwithstanding the static correlations described above, low-quality jobs 1      can be a good means of reintegrating into the labour force individuals who have poor or outdated skills. At the same time, however, these individuals will be at disproportionate risk of job loss and of being caught in vicious cycle of unemployment/inactivity and low-quality employment. It is therefore important to understand the impact on overall employment performance f:;     of better access to the labour market for the unemployed and inactive, and quality improvements for those in low-quality jobs ('upward quality ; dynamics').
52  fob Quality in European Labour Markets
Frank Siebern-Thomas S3
To this and a dynamic simulation study was conducted to assess the ( impact of labour market transition patterns on the future evolution of key indicators such as activity rate, employment rate and unemployment rate. The starting point was a (stylized) observation of labour market transitions between inactivity, unemployment and low- and high-quality employment in the age group 25-54 (Table 2.2).17
Applying this transition pattern over a 10-year period to a labour market where 70 per cent of people of working age were active (60 per cent employed and 10 per cent unemployed) and there was a 75/25 per cent split between lrigh-and low-quality jobs1" would result in a somewhat stagnating labour market. While the activity rate would remain almost constant, the employment rate would only increase by 1.5 percentage points, and consequently the unemployment rate would fall by slightly more than one percentage point.
The second step was to analyze the employment impact of different scenarios, based on different transition patterns observed in the EU member states. These scenarios varied from that in Table 2.2 only in the rates of transition into and out of low-quality jobs. The transition rates from low- to high-quality jobs, for instance, ranged from 20 per cent and 40 per cent, those from low-quality jobs into unemployment from 7.5 per cent to 17.5 per cent, and those from unemployment into low-quality employment from 15 per cent to 25 per cent.
Three major findings were obtained. First, in the absence of further quality improvements, employment creation will remain below its potential and employment rate targets such as the Lisbon target of 70 per cent by 2010 might not be achieved. With quality improvements, not only would the employment rate increase from 60 per cent to more than 64 per cent over 10 years, but also job quality, as measured by the share of workers in high-quality jobs, would increase. Second, in countries with less favourable transition patterns the activity and employment rates could decline and the unemployment rate increase. Third, strong and sustainable increases in labour market participation are required to bring up the employment rate, thus necessitating a better supply of good-quality jobs to attract inactive people back to the labour market.
Table 2,2 Annual transitions between labour market statuses, by jab quality (per cent)
Status at tStatus at (t-1)	Inactivity	Unemployment	Low-quality job	High-quality job
Inactivity	87.5	5.0	5.0	2.5
Unemployment	17.5	52.5	20.0	10.0
Low-quality job	7.5	12.5	50.0	30.0
High-quaiity job	2.5	2.5 .	5.0	90.0
Nute. The assumed labour market transition rales are based on the observed one-year transition rates in the main aye group (25-5-4} in the European Community Household Pane), waves 4-5 (1997-98).
Summary and conclusions
This chapter has laid out the policy background for quality of work and provided a summary analysis of job satisfaction, quality of work and quality dynamics in the EU, based on previous work by the European Commission. It has produced evidence of synergies between quality and quantity in the EU member state labour markets, as well as close links between quality of work and the two other principal objectives of the European employment strategy: full employment and social inclusion.
It has also shown that, while a majority of workplaces in Europe are of quite good quality, up to a quarter of workers remain in jobs of low quality, with low pay and productivity, poor training and career prospects and job insecurity, with workers being at a disproportionate risk of unemployment or social exclusion. Moreover moves from unemployment to (low-quality) employment alone, without further improvements in the quality of jobs, are unlikely to result in sustainable employment creation and increased employment rates, which puts into question some of the rhetoric in the field.19
Finally, the chapter has shown that while many of the new jobs in the service sector are of high quality, the polarization of job quality in that sector is more pronounced than in industry, with a significant numbers of employees in the retail trade, hotels, restaurants, health and social services holding low-paid or dead-end jobs. Given that most job creation will take place in the service sector, this raises questions about the incentive and labour supply effects of low job quality in some parts of the service sector.
For these reasons a broad political consensus seems to have emerged that improvements in the quality of jobs in general and those in the service sector in particular will be beneficial not only to workers' well-being, but also to overall employment performance, productivity and competitiveness, and therefore they must to go hand in hand with new policies to improve the functioning of the European labour markets.20 Such improvements are required in a range of areas, notably training and lifelong learning, gender equality, health and safety at work, work organization and work-life balance, and worker involvement. Various member states have already launched policy initiatives aimed at improving the quality of work,2' and there is an annual competition among European employers to be listed as one of the 'Best Workplaces in Europe'.22
Appendix 2.1
Data sources on job quality in the EU
The empirical findings reported in this chapter and in the European Commission's 'Employment in Europe' reports are based on data from the European Community Household Panel (ECHP, 1994-2001), the European
54  lob Quality in European Labour Markets
Foundation's surveys on working conditions (1990,1995, 2000), the European Statistics on Accidents at Work (ESAW), Eurostat's health and safety database (HASTE, Eurostat Key Data on Health 2000) and the 1995-2003 Community Labour Force Surveys, including ad hoc modules on accidents at work and occupational diseases in 1999 and working time arrangements in 2002.
The ECHP data include information on individuals and households for the years 1994-2001, with more than 120 000 observations per year. There is information on labour market status, recent job changes, objective job characteristics such as earnings, contract type, working time, job status, employer-provided training, job search behaviour, overall job satisfaction and satisfaction with specific aspects such as remuneration, job content, working conditions, job security, working hours and working time. The data can also be used to analyze transitions in the labour market.
The European Foundation's surveys on working conditions cover various aspects of job quality, such as physical working conditions (noise, extreme temperatures, repetitive tasks and so on), plus subjective evaluations of work-related health problems (fatigue, stress, backache, muscular pains), job satisfaction and sickness leave.
The ESAW database contains information on accidents at work, both fatal and non-fatal. Accidents are defined as events that lead to more than three days of absence, and fatal accidents are defined as leading to the death of the victim within a year of the accident. The data are part of the Eurostat Health and Safety Database (HASTE). The Eurostat publication Key Data on Health 2000 also includes information on working conditions and health status.
The Community Labour Force Surveys produce up-to-date data on job characteristics such as contractual arrangements, job types, training, occupation and sector of employment, atypical working times, number of working hours and job-seeking behaviour among those who wish to take up a new job or change their working time or hours.
Econometric analysis of the determinants of self-reported satisfaction levels
Given the ordered categorical nature of self-reported satisfaction levels (taking the values from 1 to 6 in the ECHP, with 1 denoting complete dissatisfaction and 6 complete satisfaction), ordered probit regression models provide an adequate framework for identifying the main determinants of self-reported satisfaction. The probability of observing a given satisfaction level i, i = l,2, ...,6, is a function of relevant individual, household and labour market characteristics (such as employment and status, job
.O
o d.
t3 «j
l-
-a o c
Q
q d
GO
m o d
| S.
2 n &
ii >■ o
o
d i
VI   O l-h
PQ cvl O O — O O
ddod
in irt in i—
* « N rt O
q q q q
o o o d s
h rt ^ io h
■* O n «     ^ o
q q q q i-h a q
c£ c£ d d d c5 a
l l l
■a- a
q q
d d l
r- O Cfr u in t-h o q i-h o o ddod
i
I
o d i
M m o -+ i-h
*rt< 1/3 i-O r-> O
q q t-> q o
d d d d d
i
i
u\ o o
N >D t r- O
q q — o o
d d d d d
■o o Ci
-H  I/}   r~* ,
q o t* i dad
i ;_
d d I
o o d d
o d
lO vo
•V <N
O O
d d
ao cm
o o a d
o d
o a d d
— o d d
-o
—s oj t
u      = -O
n oj t; o
ra =3 "S •—
JH    -a =
3
in  JZS I—i 1—i
X3
0j
m
u tu s w so o - Q < <     I w
=1 3 =5 5 ^ a. u aj   c -2
5 X si O
3 3 =
I   I   I   I I
O O Oi O N
n    o ^
q t— cn tn
d d d d d l
^E
> CL
tJ E
C C
— 3
2i bo
« c
E 3
CD
0j O
> a. c g
»s
c. £
>- O
Table 2.3 (Continued)
Satisfaction with
main activity Job satisfaction Specification 1 Job satisfaction Specification 2
Variables All All Men Women All Men Women
Older inactive 0.174" ______
Low-skilled unemployed 0.078" ______
Low-skilled inactive -0.172" - -
Highly skilled unemployed - ______
Highly skilled inactive 0.045* ______
fob characteristics:
Hourly wage	-	0.007**	0.007**	0.007"	0.006**	0.006**	0.005"
Temporary contract	-	-0.240**	-0.263**	-0.206"	-0.043"	-0.079**	-
Short part-time	-	-0.432**	~	-0.412**	-0.479"	-	-
Long part-time	-	-0.058**	-0.121*	-0.046**	-0.047"	-0.088**	-0.450**
Small firm	-	0.036**	0.022*	0.049"	0.045**	0.030"	0.039"
Large firm	-	-	-	-0.039"	-0.020*	-	-0.050**
lob status:							
Supervisory	-	0.229**	0.271"	0.146"	0.183**	0.216*	0.110"
Intermediate	-	0.132"	0.145"	0.115"	0.076"	0.080"	0.070"
lob quality:							
Good	-	-	-	-	0.125"	0.147"	0.107"
Reasonable	-	-	-	-	0.065"	0.081"	0.054"
Dead-end	-	-	-	-	-0.295"	-0.263"	-0.313"
Sector.							
Public sector	-	0.067"	0.053"	0.088**	0.054"	0.043**	0.075"
Agriculture V	-	-0.116"	-0.066*	-0.283"	-0.073"	-	-0.235"
Mining				-0.193**			-0.206"
Construction		-0.116**	-0.076"		-0.097**	-0.056*	
Retail and trade	-	0.068"	-	0.106"	0.070**	-	0.108"
Hotels and restaurants	-	0.060*	-	0.118"	0.073"	-	0.128"
Transport/communication	-	-0.051"	-	-0.130"	-0.046"	-	-0.132"
Financial services	-	0.041*	-	0.062"	0.037"	-	0.059*
Public administration	-	0.246**	0.159"	0.356"	0.239**	0.155"	0.354"
Gender concentration	-	-0.421**	-	-0.863**	-0.393**	-	-0.846"
Occupation:							
Legislators, managers	-	0.097"	-	0.337"	0.103"	-	0.322**
Professionals	-	0.129"	0.075"	0.213"	0.132"	0.081"	0.206"
Technicians	-	0.O83"	_	0.197"	0.084**	-	0.188"
Service workers	-	-	-	~	0.036**	-	0.044**
Agricultural workers		-	-	0.305"	-	-	0.297**
Craft and related workers	-	-	-0.188**	0.160"	-	-0.168"	0.159"
Plant/machine operators	-	-	-0.207"	0.196"	-	-0.186"	0.190"
Elementary occupations	-	-0.209**	-0.219"	-0.168"	-0.183**	-0.192"	-0.151"
Gender concentration	-	-	-0.339"	0.596"	-	-0.307**	0.559"
Country-specific effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year-specific effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	412917	173 636	100 569	73 067	172 213	99 748	72 465
K	34	53	52	52	56	55	55
Wald chrifc)	49 502	13 835	8 027	6 159	14 232	8 247	6312
Prob > chi2	0.000	0.000	0.000	0.000	0.000	0.000	0.000
Log likelihood	-629 625	-251 639	-146 039	-105 302	-249 240	-144 668	-104 278
Notes: Dependent variable: self-reported job satisfaction, ranging from 1 {'completely dissatisfied') to 6 ('completely satisfied'). "Significant nt the 5 per cent leve!;,* significant at the 10 per cent level; - insignificant at the 10 per cent level; - variable not included in the estimation. Estimated ancillary cut points not presented.
Source: European Community Household Panel, waves 2-7 (1995-2000).
Table 2.4  Key employment and quality indicators, total economy, 2000 (per cent)
	EU-15	BelgS	Den.	Ger.2	ger.	Sp.	Fr.	In:	Jr.	Lax1	Nesh.	Allst	Port.	Fin.	Straf.3	UK4
Employment rate (total)	63.4	60.5	76,3	65.6	55.7	56.2	62.1	65.1	53.7	62.7	72.9	68.5	68.4	67.2	73.0	71.5
Employment rate (men)	54.1	51.5	71.6	58.1	41.2	41.2	55.2	54.0	39.6	50.1	63.5	59.6	60.5	64.2	70.9	64.8
Employment rate (women)	72.8	69.5	80.8	72.9	71.1	71.1	69.2	76.1	68.0	75.0	82.1	77.3	76.6	70.1	75.1	78.1
Employment rate (low-skilled)	50.3	43.4	62.1	55.3	48.5	51.5	46.1	47.9	44.1	53.7	59.0	48.7	66.9	50.0	55.7	51.8
Employment rate (highly skilled)	82.5	85.4	88.2	83.0	80.2	75.1	78.7	86.5	81.0	80.3	86.2	86.9	89.9	84.0	82.7	87.5
Employment rate (young people)	40.4	29.1	66.0	47.2	27.1	32.3	28.6	50.7	26.4	31.9	68.7	52.4	42.4	41.1	42.2	56.7
Employment rate (older people)	37.8	26.3	55.7	37.6	38.6	37.0	29.9	45.3	27.7	26.7	38.2	28.8	50.7	41.6	64.9	50.8
Part-time employment (% of total																
employment) of which (%):	17.4	17.4	21.4	19.1	4.3	7.9	16.8	16.6	8.7	11.2	41.0	16.7	8.1	11.8	21.3	24.0
Voluntary	59	8	45	74	29	6	61	58	23	60	72	17	22	19	46	71
Involuntary	16	22	13	12	47	23	27	16	37	7	4	11	32	36	25	10
Other reasons	25	70	41	14	24	70	12	26	41	33	24	73	46	45	30	19
Temporary contract work (% of total																
employment)'1 of which (%):	13.6	9.0	10.2	12.8	13.1	32.2	15.4	5.9	10.1	3.4	13.8	8.0	19.8	17.7	14.3	6.6
Voluntary	8	10	20	2	3	1	n.a.	41	3	10	14	7	39	27	22	31
Involuntary	34	71	41	12	72	73	n.a.	27	43	13	27	18	40	59	55	32
Other reasons of which (in %):	58	19	39	86	25	26	n.a.	33	54	78	59	75	21	14	23	38
Less than 6 months	35	48	34	16	31	62	43	20	40	31	48	25	20	49	48	31
6-12 months	26	30	20	24	33	26	33	17	26	27	30	16	21	20	24	27
More than a year	38	22	46	61	36	12	25	63	34	42	22	59	59	31	27	42
Self-employed (% of total																
employment)	14.9	16.9	7.2	10.3	43.3	16.7	9.2	18.4	26.1	6.4	14.2	18.5	26.3	12.2	5.4	11.7
Job status																
Supervisory	13	12	IS	n.a.	6	7	13	12	9	n.a.	13	10	4	14	n.a.	23
Intermediate	17	20	13	n.a.	7	17	20	13	15	n.a.	16	24	6	18	n.a.	16
Non-supervisory	70	67	72	n.a.	87	76	68	75	76	n.a.	71	66	91	68	n.a.	61
Training Access to training
provided by the firm 23 61 Participation in
continuous vocational training 40 41 Gross hourly wages
Mean wages (Euros) 12.0 n.a.
Median wages (Euros) 10.5 n.a. Incidence of low pay
Share below 75% of median 23 n.a.
Share below 66% of median 16 n.a. Job satisfaction
84   n.a.    12    27   n.a.    31    17  n.a.     66 55
53    32     15    25    46    41    26    36     41 31
19.1 18.1
16 12
13.1 12.3
23 18
6.2 5.2
25 16
8.1 6.6
23 15
12.0 10.3
25 16
12.1 10.1
27 17
9.6 8.5
17 11
n.a. n.a.
n.a. n.a.
15.5 14.1
22 14
11.0
10.1
20 13
19
17
4.6 3.1
23 11
74 n.a. n.a. 50      61 49
2.1 1.0
26 11
job quality Good or reasonable Low paid/low productive Dead-end
n.a. n.a.
n.a. n.a.
14.4
12.5
26 18
High	51	61	74	n.a.	29	48	54	59	39	n.a.	66	72	26	57	n.a.	55
Medium	41	35	24	n.a.	60	43	42	36	49	n.a.	31	26	69	38	n.a.	33
Low	8	5	2	n.a.	11	9	4	5	12	n.a.	2	2	6	4	n.a.	12
74	78	81	73	67	63	74	73	79	n.a.	77	79	71	78	n.a.	74
18	18	16	14	15	10	26	19	12	n.a.	17	16	16	16	n.a.	26
8	4	3	14	18	27	n.a.	8	9	n.a.	5	5	13	6	n.a.	n.a.
NvU-.K
1. The wage data for llelgium are unreliable.
2. There are no data on job satisfaction in germany.
3. There are no data in the ECHP for Luxembourg and Sweden.
4. There are no data an access to training in the UK.
5. Excludes the responses 'not applicable' and 'no answer'.
Sources: European Labour Eorce Survey (200Ü); European Community Household Panel, UDB version March 20Ü4, wave 7 (2000); Continuing Vocational Training Survey (CVTS) (1999/2000).
o
Table 2.5  Key empioyment and quality indicators, service sector, 2000
	EU-15	Belg.'	Den.	Ger.2	Ger.	SP.	Ft.	Ire.	It.	Lux*	Nesh.	Aust.	Port.	Fin.	Swed.3	UK1
Employment rate (total)	66.9	72.2	71.1	63.9	61.3	62.3	69.6	63.9	63.0	76.7	75.2	64.2	SS.O	66.0	72.9	73.0
Employment rate (men)	56.0	61.3	59.1	51.5	56.4	51.7	58.6	50.9	55.8	67.0	65.5	52.0	46.2	51.6	59.6	61.4
Employment rate (women)	81.5	87.1	84.6	79.6	69.2	80.7	83.1	82.6	75.1	91.5	88.2	80.2	65.7	81.7	87.2	87.3
Employment rate (young people)	66.2	65.0	74.5	63.6	63.4	57.5	67.0	62.3	53.0	80.0	78.5	58.0	49.0	65.6	74.3	77.6
Employment rate (older people)	64.9	71.9	70.5	63.7	42.3	57.7	69.0	62.1	63.8	7S.0	72.9	62.3	48.5	67.3	72.5	71.2
Part-time employment (% of total																
employment) of which(%):	22.1	21.8	26.6	24.7	4.4	10.8	21.1	22.1	10.3	13.8	46.5	21.3	8.1	14.8	25.6	30.1
-Voluntary	59	8	46	73	23	6	60	58	23	60	75	17	19	19	47	71
Involuntary	17	23	14	13	54	24	28	15	38	8	3	11	46	36	26	10
Other reasons	24	69	40	14	23	69	12	27	39	32	22	72	35	45	27	19
Temporary contract work (% of total																
employment)5 of which (%):	13.4	9.7	11.0	13.5	12.8	27.7	15.5	6.8	10.2	3.7	13.1	7.6	21.0	19.3	16.5	7.3
Voluntary	9	12	22	3	4	1	n.a.	48	3	11	15	9	40	25	23	34
Involuntary	32	69	45	12	69	73	n.a.	24	40	13	25	21	39	62	55	28
Other reasons of which (in %):	59	20	33	85	27	26	n.a.	29	57	76	60	70	21	12	21	38
Less than 6 months	33	45	36	15	26	61	35	20	34	30	43	30	21	45	47	29
6-12 months	28	33	22	25	38	27	36	18	30	26	32	20	24	21	24	27
More than a year	39	22	43	60	36	12	29	61	36	44	25	50	55	34	29	44
Self-employed (% of total																
employment)	12.7	14.5	6.6	10.7	25.7	16.7	8.2	12.0	24.2	8.4	8.7	8.7	17.2	9.6	8.9	10.0
Job status																
Supervisor)'	14	7	15	n.a.	6	7	12	13	9	n.a.	11	9	4	12	n.a.	24
Intermediate	17	13	13	n.a.	7	17	21	15	16	n.a.	16	24	8	18	n.a.	16
Non-supervisory '■	69	80	72	n.a.	86	75	67	72	75	n.a.	73	67	89	70	n.a.	60
Training Access tD training provided by the firm
Participation in continuous
vocational training Gross hourly wages
Mean wages (Euros)
Median wages (Euros) Incidence of low pay • Share below 75% of median
Share below 66% of median Job satisfaction
High
Medium
Low Job quality
Good or reasonable
Low paid/low productive
Dead-end
24
46
12.2 10.7
24 16
57
88 n.a.
44 59
n.a. n.a.
n.a. n.a.
19.1 18.6
17 10
39
12.7 12.0
13    33   n.a.    34    21   n.a. 67
16    32    52    41     39   n.a. 45
6.5 5.5
25 23 19 15
57 25
75
37    26 57
8.5 7.0
23 16
12.3 12.4
10.6 10.0
25 29
17 19
10.2 9.0
n.a. n.a.
12 n.a.
7 n.a.
15.8 14.3
23 15
11.0 10.0
21 13
5.6
3.7
19 11
12.1 10.8
28 17
n.a. n.a.
59 50
n.a. n.a.
n.a. n.a.
14.3 12.3
28 20
53	n.a.	76	n.a.	35	50	56	60	43	n.a.	66	74	34	60	n.a.	55
39	n.a.	22	n.a.	57	41	40	35	47	n.a.	31	24	61	35	n.a.	33
8	n.a.	2	n.a.	8	9	5	5	10	n.a.	2	2	5	5	n.a.	12
73	n.a.	84	69	69	67	74	70	84	n.a.	77	78	75	75	n.a.	71
19	n.a.	15	15	14	10	26	20	9	n.a.	18	17	14	14	n.a.	29
8	n.a.	2	16	17	23	n.a.	10	8	n.a.	6	5	11	11	n.a.	n.a.
Nates:
1. The wage data lor Belgium are unreliable.
2. There are no data on job satisfaction in germany.
3. There are no data in the EC11P for LuxembDorg and Sweden.
4. There are no data an access to training in the UK.
5. Excludes the responses 'not applicable' and 'no answer',
Sources: European Labour l:oree Survey (2000); European Community Household Panel, UDB version March 2004, wave 7 (2000); Continuing Vocational Training Survey (CVTS) (1999/2000).
62  fob Quality in European Labour Markets
Frank Siebem-Tltatuas 63
characteristics in the case oE job satisfaction), as well as country- and year-specific effects:
	J		(	\ J		f       ; 1
Pr{i) = Pr						
			v	; = i J		I.          J=l J
where Pr(i) is the probability of self-reported satisfaction taking the value i, /=1,2, ...,6; X=(XVX1,...,XI) denotes the matrix of individual, household and labour market characteristics controlled for; P = (P,,p,,... ,p,) denotes the vector of related coefficient estimates; k^k,, ... ,k6 denote ancillary cut points; and i|i is the standard normal cumulative distribution function.
In the first model of the determinants of self-reported satisfaction with the main activity (Table 2.3), the explanatory variables included individual characteristics (gender, age and educational background), information on family background (marital status and the presence of children below the age of 12) and current employment status (unemployment or inactivity, with employment as the reference category). In addition, interaction effects between the above individual characteristics and employment status were included to analyze the differential impact of employment status on self-reported satisfaction levels by individual characteristics.
In the second model (Table 2.3) the above explanatory variables were supplemented by further information on job-related human capital (tenure on the job and skills acquired through training), specific job factors (hourly wage, working time, contract type, job status, firm size, sector, occupation and gender concentration by sector and occupation) and country- and year-specific effects. In a second specification of the model, information on objective job quality was added. These effects can be interpreted as the impact of a one unit change in the explanatory variable on the probability of the dependent variable - that is, the self-reported satisfaction level - taking a higher value (see also European Commission, 2002).
Notes
1. The EU data in this chapter refer to the EU-15 only.
2. The terms 'job quality', 'quality of work' and 'employment quality' are used interchangeably in this chapter.
3. See for example, Locke (3969, 1976) Mamermesh (1977), Kalleberg (1977) Freeman (1978) and Whelan (1980). For references to the policy debate see Seashore (1973) and Quinn etai. (1974).
4. See lor example Bluestone and Harrison (1986), Burtless (1990), Costrell (1990), Gittleman and Howell (1993), Houseman (1995), Tilly (1996a,1996b,1997), Farber (1997) and Hamermesh (1997). •
5. In the case of the United States, Tilly (1997) argues against a single summary measure of job quality and suggests seven dimensions of job quality: wages, fringe benefits, due process (protection from arbitrary discipline), flexibly hours, job permanence or security, upward mobility, and control over the work process.
6. As a complement, the European Foundation for the Improvement of Living and Working Conditions provides regular survey-based information on working conditions in Europe. See European Foundation for the Improvement of Living and Working Conditions (2002) and the references cited therein.
7. The respondents rated these factors on a scale of 1 to 6 (1 not satisfied, 6 fully satisfied). In the subsequent analysis, ratings of 1 and 2 were defined as dissatisfied, 3 and 4 as intermediate, and 5 and 6 as satisfied, or alternatively, 1, 2 and 3 were defined as as rather dissatisfied, and 4, S and 6 as rather satisfied.
8. These include permanent employment, fixed-term or short-term contracts and casual work with no contracts. In this section contract status will be defined as permanent when respondents classified themselves as being in permanent employment, and as temporary otherwise. Unfortunately the ECHP does not distinguish between temporary agency work and other types of temporary contract work.
9. The choice of a threshold of 75 per cent is motivated by the interpretation of (hourly) wages as an indicator of productivity. However it should be noted that this deviates from other measures used in the literature - notably two thirds of the median or half of the average, which are more commonly used when analyzing income inequality and household poverty. As can be seen from the wage data in Appendix 2.1, the indicator used here tends to yield slightly higher shares of employees in low-paid jobs than the other two indicators.
10. Due to the lack of information in the ECHP on employer-provided training and job status, for France and the UK only the two intermediate categories can be estimated, while for Germany the share of good jobs is seriously underestimated and that of dead-end jobs somewhat overestimated. As a result the EU-level shares of good jobs and dead-end jobs are underestimated, while those of jobs of reasonable quality and low pay/productivity jobs are overestimated.
11. See Note 10. See also Tables 2.4 and 2.5 in Appendix 2.1.
12. For more details see Tables 2.4 and 2.5 in Appendix 2.1. For an in-depth discussion of relative wages in Europe see European Commission (2003a).
13. For further details see European Commission (2003a), pp. 138-41.
14. For more detail see European Commission (2002), pp. 85-97, 110-12.
15. D'Addio etal. (2004) have found that the long-term employment performance of the various job quality clusters differs significantly. Not only is the transition rate out of low-quality employment clusters into unemployment up to four times higher than out of other clusters, but also transitions into inactivity are much more frequent. The strongest impact of low-quality employment on labour force attachment is text by women, with almost 20 per cent of all women in the low-quality employment cluster in 1995 moving to inactivity by 2000 - twice as many as in the other employment clusters. For an analysis of the link between job quality and social exclusion see Taylor (2002).
16. See Kernbeiss etal. (2003) for a contrasting view in the case of Austria.
17. See European Commission (2002) for a full description of the underlying scenarios and the simulation results. It should be noted that the analysis starts from observed labour market transitions in the period 1997-98. These transitions were probably more favourable than those which took place before the mid 1990s and after 2000. The scenarios presented in this section can therefore be considered as
64
fob Quality in European Labour Markets
Frank Siebeni-Tlwinas 65
relatively optimistic, since assuming less favourable transition patterns as a starting point would imply a less favourable employment evolution,
18. As discussed earlier, this labour market profile is roughly comparable to the EU average.
19. According to Layard (2003): 'The overall conclusion Is that a move from an average job to a bad job would reduce happiness by less than the move from an average job to unemployment, it follows that a bad job feels better than unemployment'
20. See OECD (2003b), which asks for a 'comprehensive strategy for more and better jobs'.
21. See for example the New Quality of Work initiative by the German government, social security organizations and social partners (www.inqa.de) and the various Workplace Development Programmes by the Finnish government (www.mol.fi/ tyke/00-03/en/index.html).
22. See www.eul00best.org for more information.
References
Bluestone, B. and B. Harrison (1986) 'The great American job machine: The proliferation
of low wage employment in the US economy', report prepared for the Joint Economic
Committee, Washington DC. Burtless, G. (ed.) (1990) A Future of Lousy lobs? (Washington, DC: Brookings Institution). Clark, A. E. (1998) Measures of jab Satisfaction, What Makes a Good fob? Evidence from
OECD Countries, Labour Market and Social Policy Occasional Paper No. 34 (Paris:
OECD)
Costrell, II. M. (1990) 'Methodology in the "job quality" debate', Industrial Relations, 29 (1), 94-110.
D'Addio, A. C, M. Rosholm and T. Eriksson (2004) 'Determinants of employment stability, career progression and labour market transitions', report prepared for the DG Employment and Social Affairs, European Commission, Brussels.
Employment Policy Foundation (1998) 'Job quality improves as expansion continues', e-mail (Washington, DC: Employment Policy Foundation, 1 July 1998).
Employment Policy Foundation (1999) 'Higher paying services sector jobs increasingly the rule, not the exception', e-mail (Washington, DC: Employment Policy Foundation, 14 October 1999).
European Commission (2001a) 'Employment in Europe 2001', in Quality in Work and
Social Exclusion (Brussels: DG Employment and Social Affairs). European Commission (2001b) 'Employment and social policies: A framework for
investing in quality', communication from the Commission to the Council, the
European Parliament, the Economic and Social Committee and the Committee of
the Regions, COM(2001) 313 final, 20 June. European Commission (2002) 'Employment in Europe 2002', in Synergies between
Quality and Quantit}' in European Labour Markets (Brussels: DG Employment and
Social Affairs). j. European Commission (2003a) 'Employment in Europe 2003', in Flexibility, Security
and Quatit)' in Work (Brussels: DG Employment and Social Affairs). European Commission (2003b) 'Improving quality in work: a review of recent
progress', communication from the Commission to the Council, the European
Parliament, the Economic and Social Committee and the Committee of the
Regions, COM(2003) 728 final, 26 November. European Council (2001) 'Indicators of Quality in Work', report by the Employment
Committee to the European Council, 14263/01, 23 November.
European Foundation for the Improvement of Living and Working Conditions (2002)
Qualit}' of Work and Employment in Europe: Issues and Challenges, Foundation Paper
No. 1 (Dublin: EFILWC). Farber, H. S. (1997) fob Creation in the United States: Good fobs or Bad?, Working Paper
No. 385 (Princeton, NJ: Princeton University, Industrial Relations Section). Freeman, R. B. (1978) 'Job satisfaction as an economic variable', American Economic
Review, 68, 135-41.
Gittleman, M. B. and D. R. Howell (1993) fob qualit}' and labor market segmentation in
the 1980s, Working Paper No. 82 (Washington, DC: Bureau of Labor Statistics, US
Department of Labor). Hamermesh, D. S. (1977) 'Economic aspects of job satisfaction', in Orley E. Ashenfeiter
and W. E. Oates (eds), Essays hi Labor Market and Population Analysis (New York:
Wiley), 53-72.
Hamermesh, D. S. (1997) Immigration and the Quality of fobs, NBER Working Paper
No. 6195 (Cambridge, Mass.: NBER). Houseman, S. N. (1995) job Growth and the Quality of fobs in the US Economy, Staff Working Paper No. 95-39 (Washington Upjohn Institute), i        ILO (1999) 'Decent work', by eighty-seventh Geneva, International Labour Organisation, j        ILO (2003) 'Measuring decent work', International Labour Review, no. 2003/2, special ! edition, 142 (2).
Kalleberg, A. L. (1977) 'Work values and job rewards: A theory of job satisfaction',
American Sociological Review, 42, 124-43. Kernbeiss, G., U. Lehner, R. Löffler, M. Prammer-Waldhör and M. Wagner-Pinter (201)3) 'Die Qualität von Arbeit und Beschäftigung, Methodischer Analyseraster und empirische Befunde für das Jahr 2001', Mimeo Vienna, i        Layard, R. (2003) 'Goad jobs/Bad jobs: Myths and Realities?', note presented for I discussion at a meeting of the EU member state employment and labour ministers,
i Paris, 29-30 September.
I Locke, E. A. (1969) 'What is job satisfaction?'. Organizational Behaviour and Human j Performance, 4, 309-36.
Locke, E. A. (1976) 'The nature and causes of job satisfaction', in M. D. Dunette (ed.), Handbook of Industrial and Organisation Psycholog)1 (New York: Hand McNally).
OECD (1994) The OECD fobs Study - Facts, Analyses, Strategies (Paris: OECD).
OECD (2001) 'Employment Outlook 2001', in The Characteristics and Qualit)' of Service \ Sector jobs (Paris: OECD),
:        OECD (2003a) Employment Outlook 2003: Towards More and Better fobs (Paris: OECD). OECD (2003b) 'Towards more and better jobs', communique presented at a meeting
of member-state employment and labour ministers, Paris, 29-30 September, Quinn, R. P., G. L. Staines and M. R. McCullogh (1974) fob Satisfaction, Is There a Trend?,
Manpower Research Monograph No. 30 (Washington DC: US Department of Labor). Seashore, S. E. (1973) 'Job satisfaction as an indicator of quality of employment', paper presented at the symposium on social indicators of the quality of working life, Canadian Department of Labour, Ottawa, (3-5 June). Tal, B. (2004) US employment qualit)' index: Assessing US job qualit)', C1BC World Markets Research Report (Toronto: CIBC, 21 June 2004). j        Taylor, M. (2002) 'Labour market transitions in the context of social exclusion: ! A study of the EU', report prepared for the DG Employment and Social Affairs,
i European Commission, Brussels.
' Tilly, c. (1996a) The Good, The Bad, and The Ugly: Good and Bad fobs in the United States j at the Millennium, (New York: Russell Sage Foundation).
66  Job Quality in European Labour Markets
Tilly, C. (1996b) Haifa Job: Bad and Good Part-Tmw Jobs in a Chansing Labor Market
(Philadelphia: Temple University Press). Tilly, C. (1997) 'Arresting the decline of good jobs in the USA?', Industrial Relations
journal, 28 (4), 269-74. Whelan, C. T. (1980) Employment Conditions and Job Satisfaction: The Distribution,
Perception and Evaluation of Job Rewards (Dublin: Economic and Social Research
Institute).
Alex Bryson, Lorenzo Cappellari and Claudio Lucifora*
Introduction
Deregulation of the employment relationship has characterized the mndioriing of European labour markets over the last few decades, particularly in the UK (OECD, 1994). This has permitted employers to institute greater labour flexibility and allowed them more discretion in employment relations, which are often regarded as a strategic necessity for maintaining organizational efficiency and price competitiveness in the face of intensifying market competition. It has also prompted concern about individuals' labour market prospects in terms of the availability of good jobs and the chance of prospering in those jobs through wage advancements and career development. Much of the literature focuses on wage levels, earnings progression and job insecurity. Research indicates that some groups of workers - notably women, youths and the less skilled - have borne much of the burden of increased labour market flexibility (ibid.; Gregory etal, 2000). There also appears to be a link between low-wage flexible employment and lower job satisfaction.
Despite considerable public concern and policy debate on these issues, for various reasons, not least data limitations, relatively little is known about the effects of employer behaviour on job satisfaction. This chapter aims to redress this by considering the following questions. To what extent do employer behaviour and work practices affect employees' job satisfaction? What do workers value most in terms of the pecuniary and non-pecuniary aspects of their job? Is there any trade-off between pay progression, job security and incentive pay? Are individual (subjective) expectations compatible with collective outcomes?
*We are grateful to seminar participants at the 2003 LoWER conference at the Catholic University in Milan. Alex Bryson would like to thank the Quentin Hogg Trust For funding his contribution to this chapter. Lorenzo Cappellari and Claudio Lucifora are grateful to the EU for funding the LoWER network under the Fifth framework Project.
Job Satisfaction and Employer Behaviour
67