In 2001 the Flemish Government and social partners agreed to increase substantially the quality of jobs in the region. The Flemish workability monitor was developed to help determine progress in implementing this policy. This survey data report reviews the methods and key findings of the workability monitor between 2004 and 2010. The ‘workability’or quality of work rate increased from 52.3% in 2004 to 54.3% in 2010. This means that over half of Flemish employees do not suffer from stress at work, have a job that motivates them and provides sufficient learning opportunities, and achieve work–life balance.
Download report [pdf, size 120 kb]
Policy background
Policy attention to the quality of work is inspired by the low employment rate in Belgium, especially among older workers. One of the basic conditions to increase labour market participation is to provide a good quality of jobs. In this area, policies are related to the Lisbon Strategy which speaks of ‘more’ but also ‘better’ jobs.
Belgium is a federalised country. Employment and other important policy areas are totally or partly directed at the regional or community level. In 2001, and with the Lisbon Strategy as inspiration, the Flemish government and the regional social partners concluded the Pact of Vilvoorde. This document contained a long-term socio-economic vision for the region, with 2010 as the deadline for the agreed targets.
In the Pact of Vilvoorde, the Flemish Government and social partners agreed to promote quality of work. Objective 4 states:
Thanks to an increase in the quality of work, the quality of the organisation of work and career quality, obtaining and retaining work will remain attractive for all in 2010. In 2010 the workability rate will be substantially higher.
Following the signing of the Pact, the Social and Economic Council of Flanders (SERV) committed itself to monitor the workability rate. A preparatory report concluded that monitoring the workability rate would require a new survey on the quality of work. Werkbaar werk or ‘workable work’ was the chosen title of this new survey.
In 2009 the Flemish social partners, together with the regional government and other civil society actors, renewed their social pact (just as like at European level) in Pact 2020. Learning from previous experience with the Pact of Vilvoorde, 20 ambitious objectives were set with clear, concrete and measurable targets. The ambition to create better jobs in Flanders was also renewed. One of the 2020 objectives is to create a better quality of work for at least 60% of the employees in Flanders.
Survey design
SERV commissioned Foundation Innovation & Work (Stichting Innovatie en Arbeid) to:
- develop a scientifically validated set of indicators on workability;
- organise a survey to measure these indicators every three years.
The first survey was conducted in 2004 (BE060601SD), with the second and third following in 2007 and 2010.
Conceptual framework
‘Workability’ (the translation of the Flemish policy term used in Dutch in ‘Werkbaar werk’) is defined in the survey as a multidimensional concept. A composite set of indicators is needed to catch this multidimensionality. Besides the more ‘traditional’ job quality aspects of stress and well-being at work, learning opportunities and work–family balance are added as the core dimensions of the Flemish workability monitor (Table 1). Learning opportunities and work–family balance are particularly seen as crucial points of workability (better jobs) to enhance employability and labour market participation (more jobs).
Table 1: Workability indicators in the Flemish workability monitor| Indicator | Description |
|---|
| Stress at work | The extent to which accumulated (mental) fatigue in relation to psychosocial workload leads to complaints of stress on the part of employees and to reduced job performance |
|---|
| Well-being at work | The extent to which employees are/remain committed or become demotivated due to the nature of the job (content) |
|---|
| Learning opportunities | The extent to which employees are able to maintain or further develop their skills through opportunities for training and daily experience at the workplace, with a view to their longer-term employability |
|---|
| Work–family balance | The extent to which the job demands at work interfere with the person’s home life |
|---|
Source: Bourdeaud’hui et al (2005, p. 151)
From a policy point of view, assessing ‘workability’ is only useful when insights are gained on the causes of detected problems. Workability in the Flemish monitor is therefore considered to be a bipolar concept based on cause–effect reasoning. It refers to the characteristic features of the job and its possible risks (six risk indicators, see Table 2) and to the effects of this job on the well-being of employees (four workability indicators, see Table 1).
The selection of risk factors is supported by scientific literature on well-being and stress at work, for example, the meta-analysis of Kompier (2002) which talks about quantitative and qualitative job demands, skill variety, autonomy and social support. ‘Working conditions’ was added to the indicator set to make a link with the traditional (prevention) approach to safety and health in the workplace.
Table 2: Risk indicators in the Flemish workability monitor| Indicator | Description |
|---|
| Workload | The level of workload arising from quantitative job demands such as work volume, pace of work and deadlines |
|---|
| Emotional load | The level of the workload due to contact-related job demands, especially contacts with clients or co-ordination tasks |
|---|
| Skill task variety | The extent to which the job content includes a variety of job responsibilities and makes use of the employees’ skills |
|---|
| Job autonomy | The extent to which employees are able to influence the planning and organisation of their own job tasks |
|---|
| Social support | The extent to which employees are effectively coached and socially supported by their direct supervisor |
|---|
| Physical working conditions | The extent to which employees are exposed to physical inconveniences in the working environment and to physical load |
|---|
Source: Bourdeaud’hui et al (2005, p. 154)
Methodology
Apart from some small adjustments, the surveys in 2007 and 2010 maintained the methodology used in 2004.
Workability indicators are measured through the use of psychometric scales. The survey questionnaire made use of modules selected from existing, validated instruments; for instance, it is highly complementary to the Dutch ARBO monitor examining ‘workload and stress’. Much of the survey is based on a Dutch questionnaire called ‘Experience and assessment of work (VBBA), which was developed in the mid-1990s and is based on internationally common theoretical models of stress and well-being at work such as the job demands–control–support model of Karasek and Theorell (1990). A new scale was developed for the ‘working conditions’ risk factor based on items selected from the European Working Conditions Survey (EWCS) 2000.
The data for the Flemish workability monitor are compiled from an individual random sample survey taken from a representative group of Flemish employees. The sample population is defined as ‘wage-earners residing in the region of Flanders’. In order to obtain sufficient precision and to be able reach conclusions by age, occupation and sector, the required size of the sample survey was calculated to be 8,000 respondents. Assuming a response ratio of 40%, the written questionnaire was sent to 20,000 people.
Various measures were taken for each survey (2004, 2007 and 2010) to encourage responses:
- a media campaign with advertisements;
- a telephone helpdesk to assist respondents;
- assurances of complete anonymity;
- design of the questionnaire limited to eight pages;
- sending two reminders.
The response rate was 60.6% in 2004, 53.3% in 2007 and 48.6% in 2010. Although the decreased response rate may require attention in the future, until now the surveys have provided an accurate and representative overview of the workability situation of employees in the Flemish labour market. Several sample reliability tests produced no evidence of systematic bias in the responses obtained.
Cut-off points for the workability indicators are defined on the basis of the risk effect results. Based on content-related arguments and with the help of statistical techniques, responses on the obtained measurement scales are classified as ‘unproblematic’, ‘problematic’ or ‘acute problematic’ (that is, having as a result of the score on this scale a strong or definite chance of being confronted with health or well-being problems) (see Table 3). A workable job is considered ‘unproblematic’ when all four workability indicators are labelled as ‘unproblematic’. For more information on how these cut-off points are decided, see Bourdeaud’hui et al (2005, p. 158ff).
Table 3: Components used to determine problematic workability situations| | Problematic | Acute problematic |
|---|
| Stress at work | Level of self-perceived stress-at-work that creates abnormal difficulties to recover from a day’s work | Difficulties to concentrate after work; At the end of the day problems of coping with the job due to exhaustion |
|---|
| Well-being at work | Very low intrinsic work motivation; job not perceived as attractive | High level of demotivation which has a clear effect on job performance |
|---|
| Learning opportunities | Insufficient possibilities to learn new things or for personal development | Complete lack of formal or intrinsic learning opportunities |
|---|
| Work-family balance | Systematic difficulties in coping with demands at home due to work requirements | Permanent difficulties in coping with demands at home due to work requirements |
|---|
Note: For further details, see Bourdeaud’hui and Vanderhaeghe (2010).
Survey results
Workability rate
Just over half the jobs (54.3%) in Flanders were defined in 2010 as workable by the benchmark survey, scoring ‘non-problematic’ in terms of stress at work, job motivation, learning opportunities and work-life-balance. The proportion of workable jobs grew by two percentage points between 2004 and 2007 but remained almost unchanged between 2007 and 2010 (Figure 1).
Combining the results on the four workability indicators, it can be seen that in 2010:
- 25% of employees experienced one problematic workability factor;
- 13.4% experienced two pressure points;
- 7.6% experienced 3–4 pressure points.
Occupational category
The highest workability rate was found among care and teaching professionals (59.8%) and the lowest rate (38.5%) among unskilled blue-collar workers.
In 2010 the highest workability rate was found in the construction (62.4%) and public administration (61.4%) sectors and the lowest in the post and telecommunications sector (39.3%) (Table 4). In general, workability rates were lower in the commercial service sectors.
Between 2004 and 2010 the workability situation improved in particular in the construction and health care sectors. The workability rate in the financial sector seems to have followed the ‘boom’ and ‘bust’ of this sector.
Due to the small sample size of the textile and clothing and post and telecommunications sectors in 2010, these trends should be interpreted with caution.
Table 4: Workability rate by sector| | 2004 | 2007 | 2010 |
|---|
| Total | 52.3 | 54.1 | 54.3 |
|---|
| Textile and clothing* | 43.0 | 44.7 | 51.8 |
|---|
| Metal | 52.0 | 50.9 | 50.4 |
|---|
| Construction | 51.1 | 57.9 | 62.4 |
|---|
| Food industry | 47.6 | 45.0 | 43.9 |
|---|
| Chemicals | 53.8 | 57.8 | 54.9 |
|---|
| Retail | 45.6 | 50.9 | 49.8 |
|---|
| Catering | 43.4 | 45.1 | 44.3 |
|---|
| Business services | 44.6 | 49.8 | 42.7 |
|---|
| Transport | 48.9 | 43.2 | 49.5 |
|---|
| Post and telecommunications* | 46.7 | 44.5 | 39.3 |
|---|
| Financial sector | 51.4 | 60.7 | 56.3 |
|---|
| Public administration | 59.1 | 60.3 | 61.4 |
|---|
| Health care and social work | 56.1 | 58.6 | 60.0 |
|---|
| Education | 57.2 | 59.7 | 57.8 |
|---|
Note: * Less than 200 respondents
Source: Foundation Innovation & Work
Workability indicators
Some of the main findings for 2004–2010 are presented below for each of the four workability indicators. Stress remains the main workability problem. Learning opportunities comes second, but shows the most improvement (Table 5).
Table 5: Trends in workability indicators| | % Problematic of total | % Acute problematic of total |
|---|
| 2004 | 2007 | 2010 | 2004 | 2007 | 2010 |
|---|
| Stress | 28.9 | 28.8 | 29.8 | 10.2 | 9.7 | 9.4 |
|---|
| Well-being at work | 18.7 | 18.1 | 16.6 | 8.0 | 7.9 | 7.5 |
|---|
| Learning opportunities | 22.6 | 19.9 | 18.2 | 8.4 | 7.3 | 6.6 |
|---|
| Work–life balance | 11.8 | 10.8 | 10.6 | 3.1 | 2.4 | 2.3 |
|---|
Source: Foundation Innovation & Work
More learning opportunities
The figures for learning opportunities indicate the degree to which employees are able to maintain or further develop their skills through opportunities for training and on-the-job training, with a view to strengthen their employability.
In 2010, 81.8% of the jobs can be labelled as ‘non-problematic’ for this indicator. However, this figure hides important differences by occupational category, with 96% of managers and professional workers indicating they have sufficient learning opportunities compared to only 52.9% of unskilled blue-collar workers for the same indicator. However, the situation has improved considerably in recent years for these lower-skilled workers (and not for higher-skilled workers).
Between 2004 and 2010 the proportion of ‘learning jobs’ increased by 4.4 percentage points from 77.4% to 81.8%. The sector with the highest proportion of ‘learning jobs’ in 2010 was education (94.1%); the lowest proportion of ‘learning jobs’ was found in the hotel/restaurant/bar sector (65%). Between 2004 and 2010, scores improved significantly for ‘construction’, ‘retail’, ‘banks’, ‘government’, ‘education’ and ‘government’.
Figure 2 shows sectoral differences in those learning opportunities considered ‘problematic’ for the three survey years. There was a significant change between 2004 and 2010 for routine white-collar and unskilled blue-collar workers.
Increase in job motivation increased, but not across the board
Between 2004 and 2010, the group of motivated workers (that is, well-being considered ‘non-problematic’) increased by 2.1 percentage points from 81.3% to 83.4%.
In 2010, 23.4% of blue-collar jobs could be labelled ‘problematic’ in terms of motivation compared with 18.3% of routine white-collar jobs, 8% of care and teaching jobs, and 11.6% of professional and managerial jobs.
Between 2004 and 2010, there was significant improvement in this indicator in the ‘construction’, ‘government’ and ‘education’ sectors. In the post and telecommunications sector, however, the situation deteriorated: the proportion of demotivating jobs (that is, well-being considered problematic) increased from 21.3% to 29.7%.
Better work–life balance for part-time workers
In 2004 11.8% of the workforce had problems keeping their work and family life in balance. By 2010 this proportion had diminished slightly to 10.6% though this change took place between 2004 and 2007. Between 2007 and 2010 the situation remained on average unchanged.
Work–life balance problems are mostly found in professional and managerial jobs (17.4% problematic). Between 2004 and 2010 there was a significant improvement for the construction, chemical and care sectors.
Part-time workers experience fewer problems combining work and private life. In addition, most of the general improvement in this indicator is in this group. Figure 3 shows the trends for ‘problematic’ part-time and full-time work. There was a significant change between 2004 and 2010 for those working part-time for more than 60% of the working week.
Stress at work still most important problem
Almost one in three employees (29.8%) defined their job as stressful in 2010, a position more or less unchanged since 2004 when the corresponding figure was 28.9%.
The highest problematic score is found among professionals and managers (34.6%). However, more than one in four jobs from all the other occupational categories is self-assessed by employees as stressful. The most problematic sector is post and telecommunications (39%) and the least problematic is construction (24.9%). Construction moreover is the only sector in which a significant improvement was noticed between 2004 and 2010.
The increase in stress levels in recent years has tended to be higher in the younger age groups (Figure 4).
Job risk factors
When looking at the six risk factors measured in the workability monitor, a high proportion of jobs are characterised by high workload – either in terms of quantitative job demands or in terms of emotional demands. Almost one in three jobs (30.9%) in 2010 were ‘problematic’ in terms of workload and 20.4% were ‘problematic’ in terms of emotional load (Table 6). There was no improvement between 2004 and 2010. This explains to a large extent why so many workers suffer from stress at work and why the growth in workability is so weak.
No progress was made between 2004 and 2010 in terms of task variety, with 22.8% of the jobs in Flanders being labelled as ‘problematic’ in 2010 (Table 6). This might partly explain why little progress has been made in terms of learning opportunities and motivation (that is, well-being at work). The small but significant progress (+ 2.1 percentage points in motivated workers and+ 4.4 percentage points in learning jobs) can instead be linked to the increased job autonomy (from 79.2% in 2004 to 80.5% in 2010) and to the fact that more workers received social support in 2010 from their direct supervisor(s). The rate of ‘unproblematic’ situations of social support increased from 83.9% to 85.2%.
The proportion of jobs with physically demanding working conditions did not decline between 2004 and 2010 (Table 6), with 13% of the jobs being labelled ‘problematic’ in 2010. For skilled blue-collars, there is even a negative trend with the proportion of ‘problematic’ jobs (in terms of physically demanding working conditions) rising from 26.0% in 2004 to 32.1% in 2010.
Table 6: Job risk indicators| | % Problematic of total | % Acute problematic of total |
|---|
| 2004 | 2007 | 2010 | 2004 | 2007 | 2010 |
|---|
| Workload | 31.0 | 30.6 | 30.9 | 12.8 | 12.5 | 12.3 |
|---|
| Emotional load | 20.5 | 20.7 | 20.4 | 4.0 | 4.3 | 4.1 |
|---|
| Skill or task variety | 23.8 | 23.5 | 22.8 | 11.5 | 10.5 | 10.5 |
|---|
| Autonomy | 20.8 | 20.2 | 19.5 | 7.3 | 7.5 | 6.5 |
|---|
| Social support | 15.1 | 16.0 | 14.8 | 6.1 | 6.2 | 5.4 |
|---|
| Physical working conditions | 12.1 | 13.7 | 13.0 | 4.5 | 5.3 | 4.4 |
|---|
Source: Foundation Innovation & Work
Linking risk factors and workability indicators
Although statistical analysis shows that all six risk factors increase the odds of a problematic stress situation, workload is the most important risk factor. Other important risk factors are emotional workload and lack of social support. Insufficient support and low task variation are the most important risk factors for well-being at work. Learning opportunities are specifically related to task variation, autonomy and support. Workload is the main determinant of the work–home balance.
Commentary
From a scientific perspective, it is important that Flanders has a regular benchmark survey on working conditions. But although representative data are available, the emphasis is on job content. A panel or multi-level perspective with considerable organisational context data is missing.
In their book, Working in Flanders: Exhausting or pleasant?, De Witte et al (2010) criticised the Flemish workability monitor for focusing too much on negative job aspects and risk factors and not taking into account enough and Vanderhaeghe the fact that negative job factors can be balanced by positive job factors. A final, minor comment is the slowly diminishing response rate.
From a policy perspective, it can be concluded that small improvements have been seen in the quality of work in Flanders since the social partners established the workability monitor. However, the target of a 60% workability rate by 2020 remains ambitious and specific integrated and coordinated measures are required. The workability monitor contains very useful and relevant information on ‘who’ is confronted with ‘what’ kind of problems but the design has limitations in trying to grasp what determines the occupational or sector-based profile of workability.
Guy Van Gyes HIVA-K.U.Leuven
References
Bourdeaud’hui, R. and Vanderhaeghe, S. (2010), Methodologische nota. Vlaamse werkbaarheidsmonitor (7.84Mb PDF), Stichting Innovative & Arbeid, Brussels.
Bourdeaud’hui, R., Janssens, F. and Vanderhaeghe, S. (2005), ‘Workable jobs in Flanders: the Flemish workability monitor provides a picture of quality of work in Flanders’, in Jans, M. and van der Hallen, P. (eds), Labour market research and policy making in Flanders, Garant, Antwerp.
De Witte, H., Vets, C. and Notelaers, G. (2010), Werken in Vlaanderen: vermoeiend of plezierig? [Working in Flanders: Exhausting or pleasant?], Acco, Leuven.
Karasek, R. and Theorell, T. (1990), Healthy work: stress, productivity and the reconstruction of working life, Basic Books, New York.
Kompier, M. (2002), ‘Job design and well-being’, in Schabracq, M., Winnubst, J. and Cooper, C. (eds), Handbook of work and health psychology, 2nd edn, Wiley, Chichester, pp. 429–454.
Further information
Further information on the survey (in Dutch) can be obtained from the following website:
http://www.werkbaarwerk.be
You can subscribe to an electronic newsletter which presents detailed results of the survey. A similar survey has also been conducted for self-employed people since 2007.
