Impact of the crisis on working conditions in Europe

Introduction

Context

Since late 2008 a global recession has affected the entire European economy to a greater detriment in some countries than in others. This major crisis is characterised by various systemic imbalances and was triggered by the outbreak of the banking crisis of 2007–2008. After a slight recovery in 2010–2011, the European crisis continues. This economic crisis and the related impact on employment and the labour market are changing the world of work in Europe compared with the situation a few years ago. Although not all European countries have experienced the same level of economic downturn or state financial problems, the crisis is having consequences for European working conditions.

Aims and focus

This comparative analytical report aims to show how the crisis is affecting working conditions at the level of individual EU Member States, mapping the impact of the crisis on working conditions in the broad sense.

Since the publication of its Foundation Paper No. 1 on quality of work and employment in Europe in February 2002, Eurofound has used a quality framework of employment and working conditions that includes the following dimensions:

• ensuring career and employment security;
• maintaining the health and well-being of workers;
• developing skills and competencies;
• reconciling work-life balance.

The framework relates to areas of an employee’s terms and conditions of employment such as:

• organisation of work and work activities;
• pay and rewards;
• training, skills and employability;
• health, safety and well-being;
• working time and work–life balance.

The report is not based on empirical fieldwork on these matters, but on the scanning, comparison and interpretation of existing data material and research carried out at national or European level.

The report covers the 27 EU Member States and Norway, and is based on the expert input of the European Working Conditions Observatory (EWCO) national correspondents to a structured questionnaire about the issue sent to respondents in autumn 2012. The scope of the study is therefore the period from 2008 to mid-2012. The insights obtained from the survey are complemented with available information from international studies and agencies. Findings from Eurofound’s European Working Conditions Surveys (EWCS), European Quality of Life Surveys (EQLS), the European Social Survey (ESS) and EU Labour Force Surveys (EU LFS) are used as additional comparative information when necessary. However, the picture provided is not systematic and exhaustive, but limited to synthesising the material available.

The goal of the exercise was to:

• identify the main crisis-related developments that have an impact on working conditions;
• disentangle to what extent and how changes in the economic environment, labour reforms and austerity measures have an impact in the current crisis on working conditions in Europe.

This goal was probably too ambitious. Most of the material provided on the crisis period is still very descriptive in nature and data coverage beyond the initial crisis period (2008–2010) is still limited. Assessment of the impact of labour reforms and austerity measures is ongoing. Nevertheless, these specific impacts are a point of attention throughout the text and are discussed again in the conclusion where a key approach is to compare countries that are more or less confronted with the (employment) crisis. The perspectives of the social partners also play a role in the concluding discussion.

Structure of the report

After a brief mapping of the job crisis caused by the economic crisis, the study first turns to aspects of career and employment security. Employment status, job insecurity and wage developments are briefly discussed. The career issue is tackled by focusing on training, job mobility and labour migration. Next, the report deals with elements of the reconciliation of working and non-working life. Trends in working hours, unsocial working hours, part-time work and work–family balance are described. The next section summarises trends in work intensity and organisation. In a final section, impacts on health and well-being dimensions are investigated including health problems (sickness, accidents and absenteeism) and job satisfaction.

The country information is compared by using the extent of the crisis within a country as an important context variable. Countries have suffered differently in terms of the size of the economic crisis and related policy reforms; specific policy reforms or austerity measures related to the economic crisis are illustrated at relevant places in the text.

This report is part of a wider Eurofound project on the impacts of the crisis on industrial relations and working conditions, comprising the present report based on contributions from the European Working Conditions Observatory (EWCO), a report from the European Industrial Relations Observatory (EIRO), a literature review and an overview report drawing on the two reports and the literature review. As such, limited attention is paid to topics such as wages, training and skills that are being, or have been, dealt with by other comparative analytical reports. The public sector is explicitly not a focus for this report as it will be dealt with by a separate report on working conditions in central public administration in Europe (TN1303013S).

Job crisis in Europe

The European Union has been confronted with a wave of crises affecting the economy and labour market since 2007. The economic recession began in 2007 with the banking crisis. Following this, and in some cases caused by the problems in the banking system, several European countries were faced with debt problems, involving the European Union in a sovereign debt crisis. In 2009, the world and Europe climbed out of the recession and there was some economic growth. However, this growth was too limited and unevenly distributed across Member States, causing new problems and creating the euro area growth crisis. This left the European Union facing three interlinked financial and economic crises to overcome as described in the report Euro area labour markets and the crisis (4.7MB PDF) from the European Central Bank (ECB) and the report Labour market developments in Europe 2012 (4.3MB PDF) from the Directorate-General for Economic and Financial Affairs (DG ECFIN) of the European Commission.

Employment and unemployment

Employment in Europe was severely hit by the crisis and has dropped considerably since 2009. More than five million jobs were lost in 2010, compared with the level in 2008. There was a small increase in employment in 2011, making the balance of job losses at 4.3 million since 2008.

This trend can also be found in the unemployment rates, which showed an increase for the whole EU in 2009. This increase continued in 2010 for almost all countries. In 2011 the tide turned for half the EU Member States and Norway, when unemployment rates started to decline again. Only three countries (Austria, Belgium and Germany) had a lower unemployment rate in 2011 than in 2007. Six countries (Estonia, Greece, Ireland, Latvia, Lithuania and Spain) saw an increase of more than 8% in their unemployment rates over this period (Figure 1).

Youth unemployment

The first period of the crisis had a major impact on youth unemployment (that is, people less than 25 years-old), with a rise in 2009 for all countries apart from Germany. Subsequently, part of Europe experienced a reduction in youth unemployment, together with economic recovery in 2010 and 2011. However, youth unemployment stayed above the pre-crisis level with the exception of Austria, Belgium, Germany and Malta (Figure 2). In other countries, mostly in the southeast and south of Europe, the upward trend continued.

The term ‘NEET’ is used to describe young people who are not engaged in any form of employment, education or training. The term has entered the policy debate in recent years due to the disproportionate job impact of the recession on young people (under 30 years-old). Some 14 million young people are not in employment, education or training across the EU as a whole. However, rates vary widely from around 5.5% of 15–24 year-olds in the Netherlands to 22.7% in Italy. Those with low levels of educational attainment are three times more likely to be a NEET than those who attained third-level education. The risk is 70% higher for young people from an immigrant background than nationals, while having a disability or health issue is also a strong risk factor. For further background information see the Eurofound report, Young people not in employment, education or training: Characteristics, costs and policy responses in Europe.

The unemployment rates for older workers (+50 years) show they have been less affected by the crisis (Figure 2). Nevertheless, these rates also increased considerably in those countries hit strongly by the crisis. In combination with the increased development of active ageing policies and increases in the retirement age, the policy of ‘keeping these older workers in work’ has had a moderating effect on the impact of the crisis for the employment of this group. These reforms are not dealt with in this CAR; see the Eurofound report, Role of governments and social partners in keeping older workers in the labour market, for more information.

Country clusters of economic effect of the crisis

The crisis has had a differentiated impact on the EU countries, as also indicated by general comments in the national contributions prepared for this report. Malta and Poland report being hardly hit by the crisis. Other countries (Estonia, Greece, Hungary, Ireland, Latvia, Lithuania, Romania and Spain), especially those in southern and eastern Europe together with Ireland, report a large effect from the crisis, with often a tremendous impact on unemployment, youth unemployment and long-term unemployment. There is also some variation in the duration of the crisis and its impact.

In addition, some national contributions note that the impact of the crisis on employment and working conditions seemed to lag behind the economic crisis. Nevertheless, the crisis had a significant impact on the labour market of all Member States in 2009 and 2010. The timid recovery of employment in 2011 was broken off due to the return of very low or negative economic growth rates. In the following sections of the report, this evolving crisis is considered in relation to the situation of working conditions in the EU.

To simplify and clarify further analysis of the impact of the crisis on working conditions, a preliminary classification was made of the size effect of the crisis on the EU Member States and Norway. Growth in gross domestic product (GDP) between 2007 and 2011 and the change in the unemployment rate between the end of 2007 and the end of 2011 were taken as indicators to produce a ranking of all the countries from 1 (less hit by the crisis) to 28 (strongest hit by the crisis). The countries were subsequently classified into seven groups based on their (standard) deviation for the average of these two rankings. Figure 3 shows the seven groups of countries. More details about the classification procedure are given in the technical annex accompanying this report (Tables A1 and A2). This is, of course, a simplified categorisation that is used only as a heuristic device to deal with the analytical question of how recent changes in working conditions can be related to the economic crisis.

In the middle of the map (Figure 3) are countries such as Austria, Germany and Poland where the economic impact of the crisis was very low. The next group contains the Nordic countries, Norway and Sweden, accompanied by neighbouring countries of the previous group, namely Belgium and Slovakia, and the positive southern outlier Malta. In a broader circle around these countries are Bulgaria, the Czech Republic, France, Luxembourg and the Netherlands where the size of the economic impact due to the crisis was lower. A more negative economic impact of the crisis is detected in a diverse group of countries such as Denmark, Hungary, Italy, Slovenia and the UK. A stronger impact of the crisis is seen mainly in southern countries and the Baltic States, and especially in Greece, Ireland, Latvia and Spain. Hungary and Romania score rather well in these rankings. However, in order to overcome fiscal and external imbalance, these countries have been supported by EU and International Monetary Fund (IMF) bail-out programmes – Hungary in 2008, Romania in 2009 and 2012 (and also Cyprus in 2013 after the period covered by this report).

Impact on employment conditions

In analysing the impact on employment conditions, the research looks at how secure employment conditions are and whether they are threatened by the crisis and related reforms. Feelings of job insecurity are discussed and trends in employment status are illustrated by focusing on temporary employment. This is followed by a brief examination of wage adjustments during the crisis. The career perspective on security is treated by focusing on job mobility, training and labour migration trends.

Job insecurity

Job security is measured in a large variety of ways across EU countries, as reported by the national contributions to this study. For example, it can be indicated by:

• job (in)security or self-perceived job security;
• perceived chance of losing one’s job within the next six months/year/...;
• perception of difficulties of finding an adequate new job;
• increased stress of losing one’s job.

Despite these differences, a general tendency of increasing job insecurity during the crisis years can be observed (Table 1). Poland is the one exception to this trend; job insecurity remained very low in Poland and no real change was noticed. National data for Finland, Malta and Slovenia show a recovery in job security in 2011, although the pre-recession level has not yet been reached.

Notes: nd = no data available.

National contributions: trend in job insecurity as described by national correspondent based on national statistics.

EQLS: % of working people who think it is very likely or quite likely they will lose their job within the next six months.

Source: National contributions to this study, EQLS 2007, EQLS 2012

The national contributions suggest reasons for this evolution of job insecurity during the crisis.

• The Spanish Survey on Quality of Life in the Workplace showed that employed persons in Spain felt less satisfied with their work stability and a lower percentage was looking for another job, probably as a consequence of high unemployment levels. The National Workplace Survey (2009) found that 33.7% of employees in 2009 felt their job security had declined in the preceding two years compared with just 4.3% of employees in 2003. In 2009, employees were also asked whether or not they agreed with the statement ‘my job is secure’; 29.9% of respondents ‘disagreed’ or ‘strongly disagreed’ that their job was secure.
• In Cyprus, the Cyprus Gender Equality Observatory (CGEO) reported in a working paper (in Greek) that women and young people were primarily affected by the crisis in terms of unemployment and job insecurity.
• In the Czech Republic, the results of the monthly survey ‘Our Society’ carried out by the Public Opinion Research Centre (CVVM) showed an increase of job security in 2011 following a decrease between 2006 and 2008. The author of the national contribution to this study suggests that this evolution ‘can be interpreted in a way that workers that maintained their jobs are rather those who are not easily replaceable and that occupy relatively higher positions’. Thus it is argued that the crisis might have led to changes in the composition of the labour force, which in turn can explain the increase in subjective job security.
• However, in the Maltese contribution, an indication is given that ‘the economic crisis may have changed people’s perceptions about job security’. Between 2008 and 2012 the standard Eurobarometer asked the question: ‘How would you judge the current situation of your personal job?’. Those evaluating their situation as ‘very bad’ increased but then declined again during this period. However, those workers evaluating their situation as ‘very good’ showed a steady decline (from 68% in autumn 2008 to 57% in spring 2012).
• In both Cyprus and Denmark, employees were questioned about their reasons for searching for a new job. In both countries, job insecurity and deteriorating working conditions appeared to play a more important role in this decision than before the crisis.

The EQLS (2003, 2007 and 2012) provides cross-country information on the evolution of job insecurity that allows a comparison of countries in a consistent way for the evolution of job insecurity between 2007 (before the crisis) and 2012 (Table 1). Based on the responses to the statement, ‘I might lose my job in the next six months’, a general increase in job insecurity is found. In most cases, this is the opposite to the downward trend visible between 2003 and 2007. Three countries faced a very large increase in job insecurity during the crisis (24.9% in Slovenia and 22.4% in both Cyprus and Greece) (Figure 4). In addition, significantly high changes in job insecurity are found in Ireland (12.5%), Latvia (12.4%), the Czech Republic (12.0%), Portugal (11.4%), Spain (10.9%) and Slovakia (10.4%). Four exceptions to the rule of increasing job insecurity can be noted, namely Belgium, Finland, Germany and Sweden, where the decrease in job insecurity was 0.4%–2.4%.

Figure 4 also shows a positive relation between the increase in the unemployment rate and the increase in job insecurity (based on the EQLS), which can also be concluded from the positive correlation between these two factors (0.45). There is also a positive correlation (0.47) between job insecurity and the effect of the crisis. Countries hit more strongly by the crisis also had a larger increase in job insecurity.

However, the relationship is not absolute. A difference can be seen between the new Member States (NMS) that joined the EU in 2004 and 2007 and the Nordic countries. Although the unemployment shock of the crisis was relatively lower in NMS such as the Czech Republic, Cyprus, Malta, Romania, Slovakia and Slovenia, there was a strong growth in feelings of job insecurity. Workers in the Nordic countries, even when confronted with a stronger increase in unemployment, were much less worried about this situation, with only limited (Denmark) or even negative growth of the job insecurity.

These changes in the legislation on employment protection (see Box 2) in Greece, Italy, Portugal, Romania and Spain might explain part of the large growth of job insecurity in these countries (Figure 4). However, the national contributions of these countries contained no empirical studies that investigated this relationship and some of the reforms are very recent.

Temporary employment

Temporary employment (that is, dependent employment of limited duration) can mainly be seen as the combination of temporary agency work and jobs with fixed-term contracts. There are large differences between EU countries in the representation of these two types within the group of temporary employees due to country-specific policies and customs. They are therefore discussed here as one phenomenon.

Three different trends can be found in the evolution of temporary employment during the crisis (Table 2).

• The first trend is the increase of temporary employment during the crisis due to increasing flexibilisation of employment conditions. This is the case in the Czech Republic, Estonia, Ireland, Latvia, Lithuania, Malta, the Netherlands and the UK. Policy reforms have been introduced in three of these countries (Czech Republic, Estonia, Lithuania).
• A further trend regarding temporary employment is one of a temporary decline followed by a slow recovery towards the pre-recession level (often not yet reached). This evolution can be explained by the cyclical sensibility of temporary employment. Temporarily employed workers are very vulnerable in times of crisis and, when crisis hits, temporary employees are often the first to be dismissed. But as noted in the ECB report Euro area labour markets and the crisis (4.7MB PDF) and the Eurofound report Industrial relations and working conditions developments in Europe 2011, they are also the first to be rehired with the first signs of economic recovery. The vulnerability of temporary workers in times of crisis was also the conclusion of a 2010 report on Austria’s economy (in German; English abstract) from the Austrian Institute of Economic Research (WIFO), which included the evolution of temporary employment in the country. This evolution can be found in Austria, Belgium, Finland, France, Italy, Luxembourg, Romania, Sweden, Slovakia and Slovenia, and to some extent also in Cyprus and Portugal.
• A final trend is a steady decline of temporary employment, as can be seen in Bulgaria, Norway, Poland and Spain. It is remarkable that two countries like Poland and Spain, with such a totally different economic impact as a result of the crisis, experience the same trend in relation to temporary employment. Both were countries with a very high number of temporary jobs before the crisis. However, Poland has experienced in recent years a relatively positive employment evolution, which led to the transition of temporary jobs into permanent jobs. Spain saw the opposite evolution: the massive reduction in employment was primarily achieved through the downsizing of temporary jobs.

Alongside these three trends, temporary employment remained stable in Denmark and Germany. In Greece and Hungary, the change in temporary employment is more diffuse and no clear trend can be identified.

Notes: Percentage of employed in temporary employment in 2007, 2009 and 2011 and percentage-point change between 2007 and 2011.

Temporary employment = percentage of temporary employees of total employees

Source: Eurostat

The major differences in the share of temporary employment across countries are also important. In Romania only 1.5% of employees had a temporary employment contract in 2011, while this share was 26.9% in Poland.

Although temporary employment has been affected by the crisis, it has happened in different ways in the various countries. However, the negative correlation (-0.67 percentage points) between the level of temporary employment in 2007 and the change of temporary employment during the crisis gives some indications (Figure 5). Even when Poland and Spain are excluded from the analysis because they appear to be outliers, the negative relation remains although the correlation declines to -0.41. Thus, higher levels of temporary employment in 2007 are more linked with a decrease or only a small increase of temporary employment during the crisis. Countries with a low level of temporary employment before the crisis, however, generally experienced a larger increase during the crisis. This linkage is strengthened because a range of these countries took policy measures to facilitate temporary employment (see Box 4).

When exploring the evolution within different groups of the population, some differences can be found. Apart from some exceptions (Latvia, Spain and Sweden), temporary employment has increased or decreased much more for the youngest age group (15–24 years-old) than for other age groups. This group is generally also much more represented in the group of temporary employed.

For example, in the Belgian national contribution, it is suggested that ‘young people find more often a temporary job, which increases the risk of returning into unemployment’. As noted in a report on the labour market (in Dutch) from the Policy Research Centre Work and Social Economy (Steunpunt WSE), this higher level of temporary employment can also have a long-term impact on the careers of these employees. When young employees have a large sequence of temporary contracts or stay in temporary employment for a long time (something that is fostered by the economic crisis), they lack the chances to build a stable career.

Similarly, women in Europe are more involved in temporary employment than men. However, the crisis had a reverse impact on the involvement of men and women in temporary employment. For men, temporary employment generally increased, while it decreased for women and thus the gender gap in temporary employed diminished to some extent (see technical annex, Table A3).

Two other issues to do with temporary employment deserve some attention. First, the reasons for temporary employment were certainly influenced by the economic crisis (Figure 6), with the response ‘could not find permanent employment’ gaining importance in many countries. The change was most marked in Ireland, followed by the Czech Republic, Slovakia, the UK and Latvia. Controversially, this reason also diminished in some countries and particularly in Austria, Belgium, Lithuania, the Netherlands and Poland. Secondly, the crisis had also a considerable impact on the transition rate from temporary employment to permanent employment. However, the direction and extent of this impact is very ambiguous with high peaks in both directions. It is possible that particular national policy reforms can be linked with these evolutions, however, no such suggestions are made in the national contributions on this topic.

Scientific evidence is generally consistent with temporary workers having jobs of poorer quality. A recent study on trends of job quality based on the EWCS by researchers at the LLAKES Centre, Institute of London, and Eurofound confirmed this insight (Green et al, 2011). Those employed on indefinite contracts have relatively high values on most of the indicators, while those employees with fixed-term or temporary contracts have lower job quality on all dimensions. Job insecurity and wages tend on average to be more problematic, while the effects of temporary employment status on health and well-being at work are a more debated issue (De Cuyper and De Witte, 2007).

The national contributions could shed no light on how temporary jobs have fared in relation to other working conditions in the crisis. Analysis of the European Social Survey (ESS) findings on atypical workers (such as temporary employees and workers with reduced hours) found, however, that the fear of job loss had not increased (but remained high) since the recession (Gash and Inanc, 2013). The study showed furthermore that temporary workers were much more likely to have experienced a pay cut.

Impact on wages

The impact of the crisis on wages up to 2010 is extensively discussed in the Eurofound report Wages and working conditions in the crisis. Some general trends can be observed, the main one being a deceleration and freezing of wages during the first part of the crisis. However, the crisis seems to have caused several employment effects which can partly explain the changes in wages. One of them is that proportionally more low-wage jobs were cut due to the crisis. This can have an augmenting effect on the average wage, independent of an effective increase in wages. Working hours also declined somewhat during the crisis (see below). This could have a negative impact on the average monthly wage although the average hourly wage stays the same or even increases slightly. Thus it is preferable not to just consider the evolution of the average hourly or annual wage because this might hide important employment effects. Other aspects such as changes in the composition of the labour force should be taken into account when carrying out a more in-depth investigation of the evolution of wages during the crisis.

The ECB discussed the evolution of wages in the euro area in its October 2012 report Euro area labour markets and the crisis (4.7MB PDF), which addresses some of the questions and suggestions made in this study. Similar findings about a possible composition effect are reported. The Cyprus national contribution provides further proof of this effect. In Cyprus a study on wages (in Greek, 615KB PDF) in 2010 by the Economics Research Centre (University of Cyprus) found a direct relationship between the increase in unemployment and the wage distribution. Unemployment increased more for those with low wages, such as unskilled workers and new labour market entrants.

Some other general findings were discussed in Eurofound’s report on wages in the crisis.

• Employers often prefer to save on labour costs using a wage cushion or variable pay rather than changing basic wages.
• The view that cutting wages is detrimental for the motivation and job engagement of employees was supported. Caution over cutting wages therefore seems to be necessary.
• The crisis has hit specific groups more harshly than others. Younger, low-skilled and low-paid employees seem to have experienced a greater decline in income (and more job insecurity).

These wage developments were accompanied in the recent period by policy interventions, especially in those countries confronted by severe austerity programmes and bailouts by the European Troika – comprised of the EU, International Monetary Fund (IMF) and ECB. The interventions include public sector wage cuts, freezes or cuts in the minimum wage and reforms of the collective bargaining system that facilitate downward wage flexibility. These issues are dealt with by the European Industrial Relations Observatory (EIRO) report Impact of the crisis on industrial relations and the forthcoming EWCO report on Working conditions in central public administration in Europe (TN1303013S).. Nevertheless, it is clear that these reforms have had a major impact on the downward wage trends in countries like Greece or Portugal, as reported by the national contributions to this study.

The Greek contribution states, for example, that the average nominal wage in 2011 was 6.4% less than in 2009, while labour productivity experienced a 1.9% fall. Therefore, the unit labour cost, which constitutes their difference, went down by 4.5%. If the same policy continues to be applied during the second half of 2012, the reduction of the unit labour cost would be considerable (about 8%). In this context, it should be noted that Greece is the only EU country with a nominal reduction in the minimum wage since the crisis (by 22%).

Conclusions

Job insecurity has been experienced everywhere since the crisis, most in countries hit stronger by the crisis and for specific social groups (young people, low-skilled workers and, depending on the country, men or women).

• Policy context makes a difference. Proactive strengthening of employability, a strong social safety net and clear rules on ‘losing a job’ mitigate insecurity. The job insecurity trend of the Nordic countries seems to prove this empirically. Loosening of employment protection legislation in countries like Greece, Italy, Portugal and Romania might, nevertheless, partially explain the large increases in insecurity. However, detailed studies are required to confirm this relationship.
• Trends in temporary employment diverge and depend on how important this type of employment was before the crisis. In the countries with a rather high start rate, there has been a decrease since the crisis.
• A series of mainly eastern European countries with a traditionally low level of temporary employment have introduced reforms to facilitate the use of temporary contracts. Since these reforms temporary employment has been on the rise in the Czech Republic, Estonia, Lithuania and Romania.
• It is to be seen whether a possible improvement of the economy will result in an increase in temporary jobs and a relative shift from permanent to temporary jobs. In any case, the number of people on a temporary contract stating they ‘could not find a permanent job’ has risen since the crisis.
• Freezing of wages has been the norm since the crisis. Cutting basic wages has not been the first strategy adopted by companies as a (first) reaction to the economic shock of 2008. In the recent crisis period, however, policy interventions especially in the countries confronted with international fiscal bailouts have been targeting wage cuts. The prime example in this regard is Greece where nominal average wages dropped 6.4% between 2009 and 2011.

Career development and mobility

Training, job mobility and migration are crucial aspects of career development. Developing skills and competence, changing jobs or finding a job abroad are all important steps that help to determine an individual’s work future and employment security. These three key levers of career development are discussed in the following sections.

Training

Data from the Labour Force Survey (Eurostat) allows a systematic comparison across European countries and between 2007 and 2011. Figure 7 shows the evolution and change in training opportunities. Countries are organised in terms of the increasing participation rate of employees in education and training (during the four weeks prior to the survey). Three trends can be found across Europe.

• On average, there was an increase in provided training during the crisis. This is the case in the Czech Republic, Denmark, Estonia, Luxembourg, Portugal and Sweden.
• A definite decrease in provided training on average during the crisis, as seen in Cyprus, Hungary, Ireland, Latvia, Poland and the UK.
• In the other European countries, no real (big) changes were found with the rate of participation in training remaining more or less stable.

There seems to be no relationship between the level of the participation rate in 2007 and the evolution of participation during the crisis. However, there are some differences in participation rates across Europe. The highest levels of participation in training are found in the Scandinavian countries, the Netherlands and the UK. This is the case in both 2007 and 2011 (although participation diminished somewhat in the UK in 2011). The lowest participation rates are seen in southern and eastern European countries in 2007 and 2011.

Based on individual training data from the ESS, Dieckhoff (2013) found a clearer general decline of 20% in the chances of ‘participating in training’ across the EU (and Norway) when taking into account the individual characteristics of employees – in other words, controlling for composition effects in the workforce. For example, having fewer low-skilled people in the workforce with their generally lower chance of training results in an upward bias in country averages like those shown in Figure 7. Furthermore, the ESS data show that the chances of following a training (as an employee) declined more when national GDP growth dropped more between 2004 and 2010 (Dieckhoff, 2013). However, Dieckhoff compared figures from 2004 and 2010 and not just the period of the crisis; in 2011, a range of countries experienced a (temporary) economic upswing.

The EWCO report Preparing for the upswing: training and qualification during the crisis offers some additional overall conclusions about the impact of the crisis on training, which are in line with those from Figure 7. Although the report’s conclusions summarised below are based only on data for 2007–2009, they are probably relevant for the whole crisis period.

• The decline in training offers might be due to a need for companies to reduce costs and save money. And since training is an easy-to-cut expense, this suffered during the crisis.
• There was also a change in the profile of the training provided during the crisis. Training was more often internal and more focused to the business than before the crisis. In addition, e-learning and blended ICT-based learning gained in importance.
• Nearly all EU Member State governments provided (often transitory) measures to support training aiming at empowering companies and employees via training and skill upgrading activities. The goal of these measures is to prepare enterprises to be ready for the new opportunities to create growth after the crisis.
• These support measures were found to often favour groups of workers who were more disadvantaged in terms of training participation.

Dieckhoff (2013) also found that the training opportunities were more equally distributed in 2010 than in 2004, polarisation having diminished in this working conditions topic.

These findings are confirmed in the national contributions of the current report. Some examples are given below.

Decrease

In Slovenia, there was a decrease between 2005 and 2011 in both the amount of funds (-36.5%) spent on vocational training and the number of enterprises (-26%) who provided training to their employees (data from the Statistical Office of the Republic of Slovenia).

Changing contours of company training

In Belgium, researchers at the Vlerick Leuven Gent Management School investigated the impact of the crisis on competency measures. They found that training, both general and in terms of career development, decreased during the crisis. However, training on functional competences was maintained. They also noted changes in the training given; more individual training and on-the-job training was provided while group training was less.

Policy efforts and diminished polarisation

Lifelong learning gained importance in Estonia during the crisis, with new measures being introduced such as improving access to training among unemployed and a training voucher system (2009).

Job mobility

Job mobility consists of three different kinds of movement within the labour market:

• job-to-job mobility;
• employment to unemployment mobility;
• unemployment to employment mobility.

Overall, job mobility as a whole seems closely linked to the evolution of employment and unemployment. In less favourable working circumstances – as in the current job crisis – job-to-job mobility decreases with employment, but also seems to recover very quickly with the first signs of economic improvement. Thus most European countries reported a decrease in job-to-job mobility during the first years of the crisis but an increase (or stabilisation) in the past few years.

According to the French national contribution, for example, the general amount of job mobility remained stable but the distribution over the different types changed. Employment to unemployment mobility rose, while job-to-job mobility and unemployment to employment mobility declined. This also translated into increasing numbers of unemployed.

The Belgian correspondent discusses the differences between age groups. Job-to-job mobility is greater for young people, which can be related to the fact that they often enter the labour market with a fixed-term contract or a job obtained via a temporary agency.

A report on labour markets in the Baltic states during the crisis from the Faculty of Economics and Business Administration at the University of Tartu used data from the Labour Force Survey to examine the job mobility of Estonian employees (Krillo and Masso, 2011). The study found a pro-cyclical relation between job mobility (across occupations) and economic recession. A decrease in job mobility between 2008 and 2010, especially for women, was also found.

Finally, the UK national contribution reported a drop in job mobility, especially voluntary job mobility.

Labour migration

The crisis is changing the migration patterns of a range of EU countries.

In the eastern EU Member States that joined in the past decade, emigration flows have picked up again after a (short) period before the crisis when they were declining. In Estonia there has been an increase in both employees considering working abroad and those preparing to work abroad. In Latvia, emigration almost doubled, while immigration declined. The main reason Latvians gave for emigration is ‘seeking a job or a better job’. In Slovakia, research on reasons for working abroad also illustrated the important role of job opportunities and working conditions. The top four reasons for working abroad were ‘high salary’, ‘better standard of living’, ‘better working conditions’ and ‘good employment opportunities’. Slovenia also faced increasing emigration rates with a large number of people leaving to work abroad; in particular, young and highly educated people previously working in the public sector emigrated. Thus Slovenia is at risk of losing high-level expertise (a ‘brain drain’) as a result of the crisis. This fear of brain drain is also expressed in the national contributions from Cyprus and Hungary. The main cause of this is a lack of high-skilled jobs for university graduates in these countries since the crisis.

In the western EU countries that were particularly hard hit by the crisis such as Ireland, Portugal and Spain, there has been a remarkable change in migration. In Ireland and Spain, for example, immigration dropped sharply especially during the first part of the crisis. Data from Spain’s National Institute of Statistics (INE) reveal that employment cuts in the country have especially affected migrant workers. While the employment of Spanish nationals fell by 14.3% between the second quarter of 2008 and the third quarter of 2012, employment of foreign workers went down by 24.9% (from 2.9 to 2.2 million). This reduction was especially noticeable among workers from Latin America (36.9%). In a second phase, emigration rates also grew in these countries. What is striking here is that the emigration flows consisted largely of foreign nationals returning or moving elsewhere. Thus, the crisis caused many immigrants of the past decade to leave these countries again. According to the Organisation for Economic Co-operation and Development (OECD) International migration outlook 2012, nationals who emigrated tended to choose countries outside Europe where the same mother tongue was spoken. This is the case, for example, in Ireland as reported by the national contribution.

Alongside outflows in some countries, other countries also experienced evolutions in the inflows of workers during the crisis. In Italy, planned and actual figures of immigration for seasonal and non-seasonal work decreased. A decline in work-related visas was also noted during the crisis years. Cyprus is another country with typically a considerable number of foreign workers but, in contrast to Italy, this inflow increased during the crisis. Given an increasing number of unemployed nationals and the growing national job crisis, this inflow is seen as problematic by the Cypriot government. This evolution has led to a change in the composition of the labour force in Cyprus, with more foreign workers and fewer nationals. Finally, Norway also experienced an increasing share of immigrants in the labour force during the crisis. However, here this trend was sector-related. In Norway, most male immigrants are active in industries that were less hit by the crisis, with the exception of eastern or central European workers who are mostly employed in construction. Similarly immigrant women in general are employed in health and social services, making them less vulnerable to the crisis.

These changing national migration patterns have had an impact on the comparative net migration rate for a series of EU countries (Figure 8). In particular, the net migration rate turned around and became negative in Ireland (2009), Greece (2010), Spain (2011) and Portugal (2011), countries that were severely hit by the crisis. Other EU countries with negative net migration rates in 2007 and 2011 are the more traditional emigration countries. In Estonia, Poland and Romania, the net migration rate remained stable and almost zero during the crisis. In Latvia and Lithuania, and to a smaller extent in Bulgaria, emigration appears to have grown since the crisis (Figure 8).

As well as the changes in the numbers of migrants coming and going, those working in foreign countries were often hit harder than the average national worker during the crisis. A reason for this impact is that immigrants, especially male workers, are more often employed than nationals in sectors like construction that were hit hard in the crisis. The impact of the crisis on immigrant women was much smaller; in general, their participation rates even increased as women are more active in the services sector, which until now has been less affected by the crisis. Young immigrants, often employed under a precarious contract type and on a part-time basis, also suffered more as did low-skilled immigrants. As noted in the OECD’s International migration outlook 2012, this type of worker is also more often employed on temporary contracts in vulnerable sectors.

Conclusions

• The incidence of training has generally been dropping throughout the crisis and the profile of company training has changed. Training is now more focused on the (technical) job and business requirements, and less on transversal competences not related to the job.
• Against the trend, nearly all EU Member States developed measures to increase training and the skills upgrading of especially the more disadvantaged in terms of training participation. This reduced the polarisation of the labour market in relation to lifelong learning.
• Fewer jobs results in less job mobility; people stick to their job and have fewer opportunities to change jobs.
• The crisis changed the migration patterns of a range of EU countries. Driven by an ambition to find a better job and standard of living abroad, emigration from a series of eastern European countries (for example, Latvia and Lithuania) has picked up again, while it remained stable in others (for example, Estonia, Poland and Romania). Those western EU countries that were hit hard by the crisis saw a remarkable change in their migration pattern; immigration dropped sharply, migrants are leaving the country, and more nationals are looking for work abroad.
• In the countries that mainly receive migrants such as Norway, the crisis hit the male migrant population more strongly than the female population because the crisis had a bigger effect in sectors such as industry and construction. As well as this particular gender effect, the young, low-skilled migrant with a temporary contract tended to be the one to lose their job.
• There are reports of the informal economy expanding again, for example, in countries with a traditionally high degree of informal economy such as Bulgaria or Latvia. An informal job means less social protection, more job insecurity and often worse working conditions.

Working time arrangements and work–life balance

Next to contract status and pay, working time is another major component of the employee relationship and working conditions. Part-time employment, actual working hours, overtime, unsocial working hours and subjective perceptions of work–life balance are the topics dealt with in this section.

Part-time employment

Both the national contributions and Eurostat statistics (see technical annex, Table A5) report a growth in part-time employment during the crisis. This upward trend can be found more strongly for involuntary part-time employment (that is, people work part time because they cannot find a full-time job) (Figure 9). Several national contributions suggest this is an indication that the rise in part-time employment can largely be attributed to a lack of full-time employment possibilities during the crisis.

There seems to be a positive relationship (correlation = 0.74) between the size of the effect of the crisis and the evolution of involuntary part-time employment. In Austria, Belgium, Germany and Malta, where the impact of the crisis was weaker, involuntary part-time employment even decreased, while large increases were seen after 2008 in those countries that were more strongly hit by the crisis such as Greece, Ireland, Latvia and Lithuania (Figure 9).

Part-time employment is traditionally higher for women and this situation remained so during the crisis. However, an increase can be found for both men and women. When looking at the evolution of involuntary part-time employment, the numbers for men always exceed those for women and a larger increase can also be found for men during the crisis (see technical annex, Table A6).

When differentiating between age groups, the following developments can be found.

• Part-time employment seemed to be traditionally higher for young workers (aged 15–24 years)
• In general, part-time employment increased more for young employees in the recent crisis period.
• In some countries there was huge growth in the part-time employment of young people, while the increase for the other age groups remained rather limited. This is the case in Ireland, Slovenia and Spain, with increases in part-time employment within the 15–24 age group of 22.2%, 11.4% and 11.0%, respectively. Large increases were also found in Luxembourg (8.3%), Portugal (8.2%), Denmark (8.1%) and Sweden (7.7%).
• In Austria, Belgium, Estonia and Malta another pattern can be found, that of an increase in part-time employment for both the youngest and oldest age groups. The latter can be attributed to efforts to increase active ageing and to maintain these older workers at work.

As summarised in Box 6, a range of EU countries adopted or expanded their short-time work schemes during the first phase of the crisis. However, statistically these schemes cannot be linked to the trends in part-time employment as the indicator on part-time employment is based on a spontaneous answer given by the respondent. Short-time work is more readily linked to a question in the Eurostat Labour Force Survey where workers indicate whether they are working less than usual due to business slack, plant stoppage or technical reasons. However, the lack of harmonisation of definitions hampers comparison between countries. The established trends on part-time employment are, as stressed in the national contributions, linked to the drop in economic activity. Nevertheless, the relative drop in involuntary part-time employment in Bulgaria can be related to a policy reform which introduced a part-time work scheme in 2009. Under the scheme, compensation for employees was limited to half the minimum wage and the possibility of vocational training was provided.

Working hours and overtime

Several indicators can be used to examine working time during the crisis. The first is actual working hours. The second is overtime, both paid and unpaid, which can give additional information about the working time of employees.

Changes in working hours during the crisis can be divided into two phases (Figure 10). During the first phase, 2007–2009, working hours decreased across all countries with the exception of Luxembourg, where there was a small increase of 0.3 hours. Average working hours decreased more strongly in a range of countries where the economic impact of the crisis was high (Estonia, Ireland and Latvia), but it also decreased strongly in Austria and Slovakia. In the second phase, 2009–2011, three different patterns can be found. For three countries (Austria, Cyprus and Luxembourg), the working hours remained stable. Working hours declined further in 13 countries but increased again in 12 others (see technical annex, Table A7). This heterogeneous evolution of working hours during the second period of the crisis (2009–2011) might be linked to the different general evolution of the crisis across the EU. In general, there are more increases on the left-hand side of Figure 10 (blue bars of countries with a lower economic impact of the crisis) than at the right-hand side. However, the relationship is not straightforward (see, for example, Estonia and Poland).

When distinguishing between groups of workers (self-employed, men and women), differences in the development of working hours can be found. Nevertheless, the general trend (with the two phases) remains present across all groups.

• Self-employed workers tend to work more hours than employees. This difference remained during the crisis, although the average working hours of the self-employed showed a much larger decrease between 2007 and 2009. The only increase in the average working hours of the self-employed can be found in Lithuania. Between 2009 and 2011, the working hours of the self-employed only increased in nine countries, while they remained stable in three countries and declined in the other 16.
• Men generally work more hours than women. This gender gap remained during the crisis but decreased (with exception of Belgium, Luxembourg and Slovenia). The working hours declined more for men than for women and seemed to recover less in the second phase. An important explanation for this different gender trend in working time can be related to the fact that industrial work, where more men work, was hit stronger by the crisis than other sectors.

Some possible explanations are suggested in the national contributions for the development of actual working hours. A first explanation can be found in the decline in overtime, which affected the number of hours worked (for example, in Finland and Malta). Another explanation can be found in the increase in part-time work (for example, in Belgium, Malta and Slovakia). With more people working less hours, there is a decline in the average working hours. However, when differentiating for part-time and full-time employment, in general a decline of working hours can be found in both groups (see technical annex, Table A8). However, changes in the working hours of the part-time employed is less straightforward than the general trend.

The evolution of overtime during the crisis can also clarify some of the impact of the crisis on working time. Although not all national correspondents reported on this, some general findings can be discussed. The majority of countries reported a decline in overtime, especially in paid overtime. This drop in paid overtime seems in some countries (Denmark, Slovenia and the UK) to be compensated for by an increase of unpaid overtime. Also, in Denmark, the growth in unpaid overtime is linked to an increase in compensatory leave. Some eastern European countries (Estonia, Latvia and Slovakia) reported an increase in overtime, but they are not countries where the rules on overtime were reformed and made less demanding (see Box 6). In addition, changes in overtime seem to be linked in some countries to specific sectors. For example, in Italy, overtime in general declined but there was a peak in the construction sector in 2009 linked to the effect of the general crisis and construction sites where work had to be finished quickly.

A number of policy reforms or anti-crisis measures had a direct impact on working hours. Some examples are given in the Box 6. See also the Eurofound ERM Report 2010, Extending flexicurity – The potential of short-time working schemes.

The comparative macroeconomic analysis presented in the OECD paper, The role of short-time work schemes during the 2008-09 recession, illustrates that the short-time work schemes reduced the output sensitivity of employment but increased that of average hours. The unemployment shock of the crisis was lower and the average working hours were more reduced in countries with the (higher) availability of short-time work schemes (taking into account the size of the impact of the crisis).

Unsocial working hours

Within the concept of unsocial working hours, five different groups can be defined that have working hours which differ from the general schedule of working between 09:00 and 17:00 during weekdays. A first group is those who work in shifts. Night workers are a second group and employees working during the evenings a third. The final two groups are those who work on Saturdays and on Sundays.

The fourth and fifth waves of the EWCS (in 2005 and 2010, respectively) identified an upward trend of working during the weekends and evening work over the past decades. These trends seemed to continue during the crisis while the other types of atypical working regimes tended to remain relatively stable during the crisis. For night work, some countries even saw a subtle decline. The major impact of the crisis on the manufacturing sector, where night work is widely used, can be linked to this decline. Overall, the crisis seems to have had no notable impact on unsocial working hours.

However, three national contributions stress the changing contours of working at unsocial hours.

• In France, the extent of shift work remained relatively stable during the crisis in industry and for qualified workers. However, a trend was detected of shift work shifting towards sectors and professionals that had not previously experienced systematic use of this form of work. Sectors affected were the tertiary sector and professionals such as managers (increase of 140%, though still relatively rare), administrative employees, and commerce and services employees.
• In Belgium, atypical working hours were more frequent in the industrial sector, especially for shift work, and evening and night work. This industrial sector is also more sensitive to economic changes and employs more men than women. This can help explain the pro-cyclical fluctuations in the overall share of men in atypical working hours.
• In Italy, there was a change in the production structure, with a decline in large heavy manufacturing (a ‘male-dominated sector’) and an increase in services (a ‘female-dominated sector’). This is reflected in the differences in unsocial working hours for men and women. While the male population showed a decline in unsocial working hours, women reported an increase in night work, shift work and Sunday work.

Due to changes in the economic structure as a result of the crisis, some countries saw a change in the average adoption of unsocial hours. However, no direct results are reported in the national contributions. The same goes for crisis-related reforms. Only in Luxembourg does the national correspondent report that negotiations had started in the retail sector about the opening hours of shops on Saturday.

Work–life balance

Eleven countries provided information for this report, using a diverse set of indicators, on trends in work–life balance since the crisis. The general conclusions (that is, did work–life balance improve or worsen) from this information are summarised in Table 3.

Five (Belgium, Finland, the Netherlands, Poland and the UK) of the countries reported an improvement in work–life balance during the crisis. In the other six countries (Austria, Germany, Ireland, Italy, Spain and Sweden), work–life balance deteriorated, though it was limited deterioration in countries like Austria, Germany and Sweden. In the countries seriously hit by the crisis (Ireland, Spain and to a lesser extent Italy), a higher percentage of workers reported being more worried about their work–life balance than before.

In Italy there was a clear deterioration of the work–life balance during the crisis although there were some differences between men and women. While this deterioration during the crisis is in line with the declining trend in work–life balance for women, it was improving for men before the crisis but declined again during the crisis.

Note: Only 11 national correspondents provided information on this aspect.

Source: National contributions (based on national statistics)

Conclusions

• Less economic activity means a decrease in average working hours. Part-time employment increases and overtime is cut.
• In addition, and more strongly in situations with a high crisis impact, there is an increase in an involuntary choice to work part time and an increase of doing unpaid overtime.
• The policies of short-time work schemes and working time accounts (expanded or installed in a range of EU countries in the first years of the crisis) had an additional impact on the reduction in the average working hours and reduced the unemployment shock.
• Minor changes can be detected in the average adoption of unsocial hours. These are mainly related to changes in the national economic structure as a result of the crisis.
• Work–life balance seems to have been only strongly under pressure in countries hit hard by the crisis such as Ireland and Spain.
• The gender gap changed in relation to working time and work–life balance, as men in a range of countries were hit more by the crisis and were thus more affected by cuts in working hours (increase in male part-time work, less overtime).
• Hungary, Lithuania and Portugal liberalised their overtime rules. Hungary and Romania extended the reference period for calculating working time, facilitating ‘longer working hours’. These reforms do not seem to have resulted in deviations from working time trends.

Work organisation and psychosocial risks

Various national contributions used a diverse set of indicators related to stress at work, work pressure and bullying/harassment at work to report on changes in job content and psychosocial risks during the crisis. This section first looks at the reported information on stress at work, a key factor of psychosocial well-being. It then summarises the reports on work pressure in relation to job demands and autonomy/control as a job resource. Both are widely recognised as determining work organisation factors of psychosocial risks (see Box 7). As a further trade-off of psychosocial risks, the analysis looks as the information from the national contributions on bullying/harassment and the social climate at the workplace.

None of the national contributions refers to a crisis-related policy measure related to these psychosocial risks and the organisation of work.

Work-related stress

Developments in stress at work during the crisis are documented in eight national contributions (Table 4). In general, they report an increase of work stress, although this cannot always be linked to the crisis. For example, in Belgium, there was an increase of ‘stress at work’, measured by the Workability Monitor (SERV (in Dutch)) between 2007 and 2010. However, this upward trend was also present between 2004 and 2007, which makes it difficult to link the trend to the crisis.

Other national correspondents did not find any direct relationships between changes in work stress and the crisis, but they do give some information on work stress itself. In Bulgaria, an increase was found in ‘complaints about stress’ as measured by the National Working Conditions Survey (NWCS) in 2005 and 2010. Comparing these surveys, an increase in ‘general fatigue’ also emerged. According to the Seventh National Survey of Working Conditions carried out by the National Institute for Safety and Hygiene at Work (INSHT), 23.9% of Spanish workers in 2011 said they felt ‘very stressed and coping with too much work’ compared with 20.3% in 2007. Respondents were asked in the Irish National Workplace Survey how often they found their job stressful. In 2003, 7% said ‘always’ and 47% said ‘often’ whereas, in 2009, 10% said ‘always’ and 49% said ‘often’.

In France, an analysis (in French, 743KB PDF) by the Directorate for Research, Studies and Statistics (Dares) of the SUMER data (Enquêtes Surveillance médicale des expositions aux risques professionnels) for 2003 and 2010 showed an increase in mental strain (see also FR1301011D). Data on the burnout score in the Netherlands Working Conditions Survey (2007 and 2010) showed it was at its highest level ever in 2010. Finland, however, was as an exception to the upward trend, reporting a decrease in stress based on data from the Work and Health Survey (2006–2011).

The UK contribution discusses the evolution of work stress based on two different sources. The psychosocial working conditions survey carried out by the Health and Safety Executive (HSE) found an increase in highly stressful jobs up to 2010 when the level stabilised and then fell. Biennial surveys by the Trades Union Congress (TUC) of health and safety representatives show an increase in the importance of stress as a main hazard in the workplace (2006 report (291KB PDF), 2008 report (1.3MB PDF) and 2010 report (210KB PDF)).

Notes: Evolution of work stress based on indicators in the national contributions.

* Question: ‘How often do you find your job stressful?’ (‘always’, ‘sometimes’, ‘hardly ever’ or ‘never’);

** % of workers reporting that their job is ‘extremely’ or ‘very’ stressful;

*** % of health and safety representatives identifying ‘stress (and overwork)’ as a hazard among the main hazards of concern to workers at their workplace.

Source: National contributions

Work intensity, workload and pace of work

The evolution of work intensity during the crisis is discussed in 13 national contributions, though the wide variety of indicators used limits comparisons (Tables 5–7). In addition, only current figures are reported in some cases and no information is provided on any change during the crisis.

A first indicator of work intensity is the pace of work (Table 5), reported by Denmark, Italy and Spain. In the first two countries, a decline is found based on the working conditions and health surveys by the National Research Centre for the Working Environment (NFA) in Denmark and ISFOL Quality of Work Surveys (QWS) in Italy. Spanish data indicate an increase.

Note: * Question: ‘How often is it necessary to work rapidly?’

Source: National contributions

Secondly, a change in workload gives information on work intensity, although this image is very scattered (Table 6). No comparison is made in the national contribution for Bulgaria but, in the NWCS 2010, 20% of the employees reported having to work at high speed all the time and 30% having to work to tight deadlines most of the time. Similarly, the Norway contribution only gives figures for 2009, which show that around 50% of the employees stated they had ‘too much to do’ at work. In Denmark, different sources contradict each other. According to the NFA, self-reported workload remained stable between 2005 and 2010, while a study on stress and bullying (in Danish, 163KB PDF) by the Association of Danish Lawyers, Economists, Political and Social Scientists (DJOEF) in 2010 found that 32% of the employees experienced an increased workload. In the Netherlands, speed of work and workload were measured in the Netherlands Working Conditions Survey (2007 and 2010) and remained stable. In Austria, a decrease in the impact (stressful) of working to tight deadlines was found between 2007 and 2011 based on data from the Work Climate Index (WCI). Finally, in the UK, the workload fluctuated considerably between 2006 and 2011, showing no trend according to data from the quarterly Employee Outlook survey conducted by the Chartered Institute of Personnel and Development (CIPD).

Notes: * Questions: ‘Tasks are unevenly distributed and pile up in periods’, ‘It is not possible to solve all tasks in time’ and ‘It is necessary to work overtime’.

** 1 = no, 3 = regularly, 2 items.

*** 1 = never, 4 = always, 4 items.

Source: National contributions

Some national contributions reported about work intensity in general, but like work load, no clear trend was found. In Germany, the Index Gute Arbeit 2010 produced by the German Confederation of Trade Unions (DGB) reports 36% of employees working with ‘high’ or ‘very high’ time pressure, and 61% with a higher work intensity due to this time pressure. In France, the work intensification appears to have grown in the tertiary sector while decreasing in all other sectors. In Latvia, minor changes were found in work intensification and the relationship with the crisis was not clear. In Sweden, surveys by AV in 2008 and 2012 found a reduction in work intensity between 2007 and 2011, while in Finland the initial decrease of work intensity was followed by an increase according to data from the 2006, 2009 and 2011 waves of the Work and Health Survey.

Notes: * See French national contribution for details.

** This scale is a combination of four items: ‘My job requires that I work very hard (2003: 82%; 2009: 89%)’; ‘I work under a great deal of pressure (2003: 51%; 2009: 58%)’; ‘I never seem to have enough time to get everything done in my job (2003: 38%; 2009: 47%)’; and ‘I often have to work extra time over and above my formal hours to get through the job or to help out (2003: 47%; 2009: 48%)’.

*** Percentage of employees who have to extend their working time due to the demands of tasks or managers.

**** Proportion experiencing they have too much work to do.

Source: National contributions

The wide variety of data on work intensity and the trends in work intensity during the crisis make it difficult to draw straightforward and clear conclusions. There is very little or no information on work intensity from those countries more strongly hit by the crisis apart from Spain, which signals an increase. However, of the countries that were less strongly hit by the crisis for which information is available, both Austria and Sweden report a decrease in the work intensity. Those countries more averagely impacted by the crisis provide a very unclear image, with some reporting increases of work intensity and others reporting decreases or stable levels of work intensity.

The Nordic countries (Denmark, Finland, Norway and Sweden), Ireland and UK document developments in work intensity in a much more detailed way than other European countries, primarily due to a greater focus on these aspects of work in the former set of countries. However, no clear trend can be seen for these countries compared with the rest of Europe.

A recent study on the ESS (2010 and previous waves) gives some more comparative information considering work stress and work intensity during the crisis (Gallie, 2013). It shows that research on these topics found a trend of work intensification since the early 1990s, with a stronger increase of work intensity during the crisis (between 2004 and 2010). The study also found some support for the view that the economic crisis led to this stronger increase: work intensity increased more in companies that experienced financial difficulties and staff reductions in the past three years. However, it was also found that this increased work intensity was not necessarily related to higher stress. Only in some countries (the Baltic states, France Ireland and the UK) did the increase in work intensification lead to more highly stressful jobs.

These findings suggest that more elements influence work stress than solely work intensification. Karasek’s job demand–control model (see Box 7) gives more clarity on this topic, especially considering the role of job control. Other research from Russell and McGinnity (2013) on the scale of work pressure in Ireland (2003 and 2009) shows evidence of a clear relationship between the increase in work pressure and the crisis in terms of increasing job insecurity, staff cuts and company reorganisations. These three factors appeared to have had a significant impact on work pressure in Ireland.

Job control

Below the JDC model is used (see also Box 7) to discuss developments in work organisation and job control to obtain a better view on the impact of the crisis on work stress.

Three national correspondents reported on changes in job control and autonomy during the crisis. In Denmark, a decrease in the employee influence on the planning and method of solving tasks was found. Similarly, in France, there was a decrease in job decision latitude. According to the available data, 20%–35% of Spanish workers do not have the option to choose to modify the working method, the working rhythm, the order of tasks or their own ideas at work; notwithstanding this, these indicators have improved since 2007.

The recent study on the ESS (2010 and previous waves) by Gallie (2013) discusses job control and its relation to work stress. Previous research has often found (in line with the Karasek model) that the impact of increasing work intensity on psychological well-being and physical health is less negative when job control is higher (active jobs). Job control is traditionally higher in Scandinavian countries and lowest in southern and eastern European countries. The study by Gallie (2013) also found that job control during the crisis predominantly increased, certainly in the southern and eastern European countries, confirming the findings of the ERM report on the EWCS data (see Box 7) and the Spanish results.

The study by Gallie has two main explanations. First, it is suggested that the selective displacement of employees – with those leaving employment being low-skilled and often having the least job control – led to an increase of the average job control. Secondly, the study finds evidence for the fact that employers tend to give more job control to employees when they have to restructure, with the aim of enhancing their commitment to the organisation. The study therefore reports an increase in both work intensity and job control during the crisis. In the light of the JDC model, this can suggest that not all the increase in work intensity during the crisis led to an increase in stress (with negative influences on both the physical and psychological health of employees).

Conflict, bullying and violence in the workplace

Extreme expressions of a deteriorating psychosocial environment at the workplace are issues such as bullying, violence and threats of violence, and (sexual) harassment. The national contributions report the following.

In the Czech Republic, a study by the Public Opinion Research Centre (CVVM) found a steady increase in cold, impersonal relationships with colleagues between 2006, 2008 and 2011. Similarly, employees were more likely to report an atmosphere of conflict at work in 2011 than in 2006 and 2008. Thus, the crisis seems to have negatively impacted the working climate in the Czech Republic. However, according to data from the Netherlands Working Conditions Survey, there was a decrease in conflicts and problems at work between 2007 and 2011 in the Netherlands.

An increase in bullying at the workplace during the crisis in Denmark has been documented by the NFA. In the UK, the TUC also found an increase between 2008 and 2010. In Norway, bullying and teasing by colleagues seems not to have been affected by the crisis, with levels remaining stable over the past 10 years according to a report from the National Institute of Occupational Health (STAMI), Facts about work and health in 2011 (in Norwegian).

Physical violence in the workplace also seems to have increased in Denmark, as do the threats of violence according to the NFA. In Poland, an increase in physical violence by male employees was found between 2009 and 2012 by the Public Opinion Research Centre (CBOS). In Portugal, physical violence also increased between 2008 and 2009 according to reports from the Authority for Working Conditions (ACT). However, there was no clear trend in violence at work in the UK between 2006 and 2007 and between 2010 and 2011, according to a HSE report based on the British Crime Survey. In Romania, a study carried out by the National Trade Union Bloc (BNS) into working conditions, satisfaction and performance at the workplace reported a decrease in both physical and mental violence between 2010 and 2011.

An increase in harassment at work was reported in Italy, Portugal and the UK, while a decrease was found in Romania. In Slovenia, sexual harassment in the workplace remains an element of concern according to the national correspondent.

Conclusions

• An increase in stress at work is reported although this cannot always be linked to the crisis.
• Scattered national data confirm the analyses of international survey data (EWCS and ESS) that higher levels of work intensity (workload, work pressure and job demands) are found in workplaces that have restructured for economic reasons, but that these higher job demands are accompanied by higher levels of autonomy and job control. However, the increase in job resources is less than the increase in work intensity, resulting in small average increases of stress at work. Spanish survey data, for example, confirm this pattern at the aggregate level: higher average work pressure and pace, but also improvement of job control figures; the rate of ‘stressed’ workers increased from 20.3% in 2007 to 23.9% in 2011.
• Reports from the Czech Republic, Denmark, Italy, Portugal and the UK suggest that conflict, bullying and violence are on the increase.
• Rising job insecurity plays an important intermediating role in the final effect on well-being at work (stress, harassment).
• No policy crisis measures in relation to the organisation of work (and its innovation) are reported.

Health and well-being at work

Finally, the study looks at what could be called the ‘outcome’ indicators of working conditions in the crisis when information on absenteeism, accidents at work, occupational diseases and job satisfaction is compared.

Absenteeism and presenteeism

Developments in absenteeism are documented in 19 national contributions based on a variety of indicators, including average number of days of absence, claims for sickness benefits and absenteeism rate (Table 8).

Notes: See Annex 1 for a full list of country codes. No data are available for Austria, Finland, France, Greece, Hungary, Ireland, Romania, Slovakia and Spain.

Source: National contributions

The most widespread trend is a decrease of absenteeism during the crisis years, which is reported in at least 11 countries. However, four countries (Belgium, Germany, Luxembourg and Norway) reported an increase. Three countries (Cyprus, Latvia and Poland) saw an increase during the first years of the crisis, with a decrease afterwards. In the Netherlands, the opposite was found.

Some national contributions present suggestions to explain the mainly downward crisis-related trend. In Malta, for example, the decrease of absenteeism is suggested to be a direct result of lower employment levels in the public sector, where absenteeism was traditionally higher. Existing academic research sheds more light on the possible explanations for the reported decrease of absenteeism. Since the occurrence of absenteeism cannot be seen apart from presenteeism (‘phenomenon that employees go to work despite feeling so ill that they should have stayed home’ (Heponiemi et al, 2010)), this crisis phenomenon also implies an increase in presenteeism.

A first group of general determinants of absenteeism are the demographic characteristics of the employee. According to a pan-European study on sickness absence from the Munich Personal RePEc Archive (MPRA), absenteeism is generally higher for women, blue-collar workers, workers with a low level of educational attainment and singles; in addition, absenteeism increases with age. Based on the crisis trends in employment, one can predict contradictory trends in absenteeism. Employment of men and young people falls during the crisis, which should lead to an average increase in absenteeism. Blue-collar, low-skilled jobs are also cut more in periods of economic crisis which would, according to the hypothesis, lead to an average decrease in absenteeism.

The report has already highlighted the considerable increase in job insecurity since the crisis. Past research has clearly established that job insecurity decreases absenteeism (and increases presenteeism) because employees fear job loss or loss of future employment opportunities when they are absent (Heponiemi et al, 2010; Prater and Smith, 2011). Moreover, when unemployment rates are higher such as during an economic crisis, job insecurity will increase presenteeism even more (Hansen and Andersen, 2008).

Institutional factors also influence the likelihood of absenteeism. When sickness benefits are generous and employment protection is high, as noted in the aforementioned MPRA report, absenteeism will be higher. However, low or no sickness benefits, often in combination with low wages or a poor personal financial situation, prompt employees to be present at work even though they are too ill (2004 report on sick leave from the Institute for Women’s Policy Research; Aronsson and Gustafsson, 2005; Dew et al, 2005). The introduction of more stringent rules for sickness benefits since the crisis in Estonia, Italy, Lithuania, Norway and Portugal are reported to have had the effect of reducing rates of sickness leave (see Box 8).

Those national contributions that report something about presenteeism confirm it was on the rise during the crisis. In Bulgaria, for example, it is suggested that the increasing presenteeism is linked to attempts to avoid a loss of income and fear of being dismissed when taking sickness leave. Also in the UK, there seems to be a relation between the increase of employees coming to work ill and planned redundancies. So however limited, the results on presenteeism described in this report seem to be in line with what other researchers suggest happens during economic recessions.

Conclusions

Rising job insecurity and in particular policy reforms seem to be important factors in explaining a declining crisis-related trend in absenteeism (sickness leave). However, those countries reporting an increase in absenteeism are faring relatively well in the current crisis. Furthermore, the Belgian correspondent relates the continued increase to long-term trends of an ageing workforce and deteriorating psychosocial working conditions (stress, burnout). The national contributions from the countries with contradicting trends (increase/decrease) provide no specific explanations for this evolution.

Occupational accidents

From the combination of the information from the national contributions and the standardised incidence rate of occupation accidents from Eurostat, a general downward trend emerges in the occurrence of occupational accidents in almost all countries, except for Ireland and Poland, during the crisis (Table 9).

Several indicators are used in the national contributions to measure occupational accidents. In general, a decline in occupational accidents is found, especially during the first part of the crisis (up until 2010). The number of days lost due to occupational accidents, when reported, also declined during the crisis. The drop in accidents seems to be more prevalent in those sectors that are traditionally more at risk such as construction, manufacturing and industry.

Some national correspondents also suggested explanations for these developments. In Spain, two factors may have contributed to the decline of occupational accidents:

• public awareness campaigns and regulations on occupational health and safety (OSH);
• the workload reduction in some industries (not overall).

This last factor can account for the strong reduction in accidents in the construction and industry sectors where the loss of employment can also further explain the situation.

Spanish survey figures show that workers now feel less exposed to accident risks. Similar explanations are suggested by the Finnish correspondent who argues that occupational accidents may have declined because:

• companies gave employees more time for training (thus reducing the risk of injuries) rather than laying them off;
• time and work pressure declined during the crisis, causing employees to work more safely and more carefully.

Also in Estonia, the decline is partly attributed to the less intensive work during the crisis. In Bulgaria, considerable attention went into to improving OSH during the crisis, with new laws and a new regulatory framework introduced. Thus, the number of companies meeting the basic OSH requirements rose, as did those having a risk assessment procedure and a programme to eliminate those risks. These changes and OSH initiatives could also explain the reduction in the number of accidents. However, the German national correspondent links the increase in occupational accidents in 2010 recorded in a joint report from the Federal Institute of Occupational Health and Safety (BAuA) and the Federal Ministry of Labour and Social Affairs (BMAS) to the economic growth and rise in employment. Although no data are given about occupational accidents in Norway, the national contribution discusses the negative influence of the crisis for employees in the construction sector concerning health and safety. The crisis caused the security to deteriorate and increased the risk of accidents. In 2009, the Labour Inspectorate found an increased number of violations of safety regulations during their national campaign in the construction sector.

The Eurostat LFS data allow comparisons across countries and time in Table 9; since the data for 2007 are limited, comparison is made between 2008 and 2010. Also no data were provided for Greece. In general, a decrease is found between 2008 and 2010, with the exception of Ireland and Poland. However, these data call for careful interpretation. While the standardised incidence rate of occupational accidents between 2008 and 2010 shows an increase only in Ireland and Poland, a different pattern emerges when looking at the change year by year. Between 2008 and 2009, occupational accidents decreased in almost all countries. However, 14 countries show an increase of the standardised occupational accidents rate between 2009 and 2010 – although this increase is smaller than the decrease the year before. This trend of a decrease followed by an increase is also reported in two national contributions (Estonia and Germany).

Notes: Eurostat standardised incidence rate is the number of people involved in accidents at work with more than three days’ absence per 100,000 people in employment.

See Annex 1 for a full list of country codes.

nd = no data available.

Source: Eurostat, national contributions

A comparison of the Eurostat standardised incidence rate for fatal occupational accidents between 2008 and 2010 is given in the technical annex, Table A9. However, determining the change in the number of fatal accidents is less straightforward. In 14 countries there is a decrease in fatal accidents whereas an increase (although sometimes very small) is noted in 10 other countries. The impact of the crisis on fatal accidents is discussed in only a few national contributions. In Denmark, 93% of all fatal accidents at work between 2006 and 2011 involved men. This can be explained by the fact that more men are employed in those sectors where the incidence of fatal accidents is high. In the Czech Republic, a sharp drop in fatal injuries was found in 2009, which could be due to the high numbers of job losses among employees working at hazardous workplaces.

Research on occupational accidents is in line with the findings from the national contributions and Eurostat, in also reporting a positive relationship between the business cycle and occupational accidents. In times of economic growth, the probability and incidence of occupational accidents increases. They then decrease when an economic downturn occurs (Asfaw et al, 2011). Explanations for this relationship can be grouped into the following categories, which often are also suggested by national correspondents.

• Changing composition – A recession causes changes in the composition of the labour force. Fewer new employees, who are more prone to accidents, enter the organisation and when downsizing occurs, the most experienced employees often remain. Thus the labour force is more experienced, diminishing the probability of occupational accidents. In addition, the crisis in many countries has had a major impact on industry and construction, sectors traditionally more prone to occupational accidents.
• Lower work intensity – Recession often has an impact on working conditions and work organisation. It is suggested that workload and pace of work tend to decline, creating more time for training workers (for example, increasing their experience working with machines) and maintenance (HSE research report on workplace injury (2.7MB PDF) published in 2005; Davies et al, 2009; Asfaw et al, 2011). Correspondents from Estonia, Finland and Spain also point to this explanation. However, as noted in the previous section, this decrease in work pressure is not always happening. The element of control also needs to be taken into account. In other words, the impact of this factor has to be interpreted with caution for each country separately.
• Underreporting – During times of higher unemployment and job insecurity, employees are less likely to report (minor) accidents because they fear this might be used against them. This fear is confirmed in a study by Boone et al (2011) who found that ‘workers who report an accident in a particular period of time are more likely to be fired later on’. Thus, when looking at a trend in occupational accidents, it is important to keep in mind that this can be due either to an actual change or to a change in reporting behaviour.
• ‘Less a priority’ as counterfactor – The Norwegian national contribution highlights a counterfactor; companies and sectors hit hard by the crisis (like construction in Norway) have more difficulty maintaining and prioritising (new) health and safety measures.

Occupational diseases

Only national information is available on trends in occupational diseases (Table 10). Thirteen countries reported on trends in occupational diseases, of which five (Bulgaria, Estonia, France, Germany and Spain) showed an increase over the period of the crisis and five (Austria, Poland, Portugal, Romania and Slovenia) showed a decrease. Three countries reported no significant change between the beginning and end of the crisis, but with some fluctuations in between. In the Netherlands, an increase was followed by a decrease, while Denmark and Luxembourg showed the opposite trend. No explanations for the observed trends were suggested in the national contributions.

Note: No data available for Belgium, the Czech Republic, Greece, Hungary, Italy, Latvia, Lithuania, Slovakia, Sweden, the UK and Norway

Source: National contributions

The following general health tendency figures from Greece, perhaps the country most struck by the economic crisis, are illuminating. The Depression Helpline of the University Mental Health Research Institute (EPIPSI) has registered a surge in anxiety and depression symptoms that are mainly due to a lack of work perspective, professional insecurity and uncertainty, failure to find work or unemployment. Six out of ten people who called the helpline for support said they were unemployed; the reason given by 17.7% of those who could not find a job was the general unemployment in Greece. According to collected data, 64.4% of these people are women, 37.3% are young people aged between 21 and 35 years, 46.7% are tertiary educated graduates, 61.6% are Attica residents (the administrative region which includes Athens) and 4.34% are married individuals. The president of the Cyprus Psychologists Association also warns of a similar relationship in Cyprus between economic problems and suicides, as is found in Greece.

Job satisfaction and motivation

Job satisfaction and job motivation/engagement are seen as essential components of well-being at work. But although these general job attitudes of well-being are closely related to each other, they are not the same. Job satisfaction refers to the satisfaction that a person receives from performing their job. Motivation is about being engaged or having the drive to fulfil a job. Motivation is about why somebody is doing a job; satisfaction is about the attitudinal evaluation of doing this job.

European comparative survey data from the EQLS and ESS highlight a general trend of increased job satisfaction. The EQLS has data for 2007 and 2010, and the ESS for 2006 and 2010. Figure 11 shows comparative data from the ESS 2006 and 2010. On a 10-point scale, the EU average of job satisfaction increased in the EQLS survey from 7.1 to 7.4 between 2007 and 2010. The report detailing the findings of the third EQLS – Quality of life in Europe: Impacts of the crisis – shows that the increase is consistent for all age, gender and income groups.

Some of the national contributions, based on specific national sources, add nuances to these general figures (Table 11).

• National data from the Czech Republic, Ireland, the Netherlands, Poland and Spain confirm the (slight) increase in job satisfaction, while the Austrian WCI data do not. In 2011, however, the Czech and Dutch figures also slightly decreased. Italian and Latvian data remained stable.
• Data from Italy, Ireland and Spain (countries badly hit by the crisis) mostly confirm the increase, but show also that pay satisfaction decreased or remained stable.
• Survey information from the Czech Republic and Poland illustrates diverging trends within these countries. In Poland, the proportion of those reporting satisfaction increased as did the proportion reporting dissatisfaction. In the Czech Republic, those reporting being ‘very satisfied’ decreased, but so did the proportion being (very) dissatisfied. The group who were undecided about their job satisfaction grew in the crisis period.

Source: National contributions

Employees thus appear more likely to be satisfied with the type of work they do when confronted with a negative business cycle. This is in contrast with the negative relationship of job insecurity with satisfaction reported in Box 3. A series of recent studies explain these contrasting results. During economic expansion, working in sectors with high unemployment tends to be associated with higher job insecurity and thus lower job satisfaction. However, during a crisis, a countervailing effect from social comparisons emerges. The negative effect of having worse employment prospects is compensated by the positive effect from their relative standing in relation to those unemployed (Clark et al, 2010; Luechiger et al, 2010).

The measures used to assess workers’ motivation differ considerably. Three national contributions reported on the impact on job motivation during the crisis. In Finland, an improvement in meaningfulness of work was found during the crisis, but this decreased again in 2011. In Ireland, the recession had a positive effect on commitment and motivation, especially in relation to the general organisation (organisational citizenship behaviour). In France, 47% of employees stated in a survey on employment beyond the crisis (in French, 820KB PDF) by TNS Sofres for the Employment Stewardship Council (COE) that they thought the crisis had impacted negatively on the motivation of their co-workers.

Steiber (2013) analysed data from the 2010 ESS and previous waves of the survey to see if there was a relationship between commitment to employment (as a measure of work motivation) and elements of the crisis (job insecurity and unemployment). Employment commitment was found to be higher in Scandinavian and western European countries (70% committed), while workers in southern and eastern European countries tended to have lower levels of commitment. The study found no support for the hypothesis that work ethic is negatively influenced by unemployment (since 2000). However, job insecurity was found to have a negative impact on employment commitment, especially in southern European countries and NMS from eastern Europe. Motivation thus seems to be mostly affected by the crisis through the increase in job insecurity.

Conclusions

• A drop in absenteeism can be discerned in most of the countries for which information is available. The rate of accidents at work also went down, especially between 2008 and 2009, but rose again between 2009 and 2010 in several countries (which showed some economic recovery within that period). Different, even contrasting, explanations are given for these trends. The fear of losing one’s job could lead to less absence and less reporting of accidents. Others highlight the fact that the crisis led to less work pressure because there was less to produce. Another explanation points to the fact that the more experienced and settled workers are kept in the job during a crisis, and these are people with a lower chance of being absent or being involved in a work accident. Detailed further research is needed, however, to show which of these arguments is the most plausible.
• Nevertheless, it is clear that more stringent rules for sickness benefits introduced since the crisis in Estonia, Italy, Lithuania, Norway and Portugal have had the effect of reducing rates of sickness leave.
• Trend information on occupational diseases is mixed. It is of concern that in a country like Greece, which is strongly confronted with the crisis and austerity reforms, there is increasing debate on the effect of the crisis on general health and the rising suicide rate.
• Although national particularities can be detected in the trend data, longitudinal comparative data from the EQLS and ESS surveys suggest that job satisfaction has risen in Europe since the crisis. The negative effect of rising job insecurity is compensated by the happiness in (still) having a job. Motivation seems to be hampered by the increased job insecurity.

Conclusions

This report has investigated the impact of the current economic crisis on working conditions in Europe based on reporting from national experts and supported by available international data and comparative studies. The focus has been on assembling national trends and findings from the 27 EU Member States and Norway. From a chronological perspective, the period of crisis is covered mainly until the first months of 2012.

Global findings

Throughout this report the following pattern emerges – rising insecurity, less choice, wage freezes and the feeling of ‘not being all in it together’. The same pattern was found by the 2011 Workplace Employment Relations Study (WERS) when UK workers were asked what kind of changes they had experienced in their job in the recession. The main changes detailed in the report were no pay increases, higher job insecurity and a higher workload. Some 40% of respondents reported no change, a minority had fewer training opportunities, while few respondents had changed jobs.

Less work

First, crisis means less work with jobs being cut. The first period of the crisis was also characterised by an overall reduction in working time and more part-time work because there was less economic activity. In a second period, there were contrasting developments in average working hours related to the extent of the crisis effect in the country.

A majority of countries also reported a decline in overtime, especially in paid overtime. Working at unsocial hours has not expanded during the crisis.

More insecurity and consequences for workers’ well-being

The economic crisis is a factor in job stress and insecurity. Concerns rise about having or keeping a job and about establishing or maintaining income. Job insecurity has increased across Europe. The negative consequences of this insecurity on well-being and health are a recognised scientific fact. Illuminating is in this regard is the rising trend in depression and anxiety mentioned in the Greek national contribution.

Basic wages are not the first costs cut by companies. However, wage freezes and cuts in the countries hit hard by the crisis and by fiscal bailout programmes have contributed to increased income insecurity.

Ongoing restructuring and downsizing are also expected to increase stress for those who stay in work. The partial evidence submitted for this report seems to confirm this expectation.

There is some evidence that harassment, bullying and mobbing are greater risk factors in this period of economic stress.

Work intensity and workload

Scattered, national data confirm the analyses of international survey data (EWCS and ESS) that higher levels of work intensity (workload, work pressure and job demands) are found in economically restructuring workplaces, but that these higher job demands are accompanied by greater levels of autonomy and job control. However, the rise in job resources is less than the rise in work intensity, resulting in small average increases of stress at work. For example, Spanish survey data confirm this pattern at the aggregate level; higher average work pressure and pace of work, but also an improvement in job control figures (the rate of ‘stressed’ workers increased by 3.6 percentage points between 2007 and 2011 to 23.9%).

Less choice

A next, very important effect of the current crisis is that it diminishes the choice possibilities of workers. Some national contributions mention that people have had to stick with their current job because no alternative is available, while others in contrast have been obliged to go and look for a new job. Atypical employment forms (part-time work or a temporary job) are chosen involuntarily because there is no other employment. This phenomenon has risen significantly since the crisis period. This ‘less choice’ situation can also be detected in other job elements. Overtime has indeed dropped, but unpaid overtime has not. Presenteeism (going to work though being ill) has grown.

Career mobility is negatively influenced by a crisis situation. Based on an analysis of micro data and less on country averages, training possibilities are seen to be somewhat lower and more directly job-related. Job-to-job mobility is much lower: people stick or have to stick to their current job.

Work–life balance seems to have particularly deteriorated for workers in the countries hard hit by the crisis (for example, Ireland and Spain).

Extreme forms of this ‘less choice’ situation are that people in some countries are again being pushed into the informal economy or having to look abroad to find a (decent) job and living. A growth in the informal economy is reported in Bulgaria, Cyprus and Latvia. Eastern European countries such as the Baltic states of Latvia and Lithuania managed just before the crisis to counter the exodus of the previous period. They are now confronted once again with emigration. Countries like Greece, Ireland and Spain have turned from net immigration countries into emigration countries.

More positive findings on other indicators

National trend averages provide more positive conclusions about health and well-being at work.

Most of the countries saw a drop in absenteeism. The standardised incident rate of accidents at work went down between 2008 and 2009, but rose again in about half of the countries between 2009 and 2010. Different, even contrasting explanations are given for these trends. The fear of losing one’s job could lead to less absence and less reporting of accidents. Others highlight the fact that the crisis leads to less work pressure because there is less to produce. Another explanation is that the more experienced and settled workers tend to be kept on and have a lower chance of being absent or being involved in a work accident. Future research should not only examine which of these arguments is the most plausible but should also look for sectoral differences as the effect of the crisis in a series of countries is linked to specific sectors such as industry and construction.

On average, job satisfaction has risen since the crisis. Work motivation is mainly under pressure when job insecurity is high.

More scars to come

It is difficult to grasp the effects of an ongoing process like the current economic crisis and accompanying policy interventions to combat it. The two main arguments for this position are as follows.

• Average trends have received considerable attention. However, these aggregates hide composition effects. For many companies the crisis has meant that so-called peripheral jobs have been cut. More low-skilled jobs than highly skilled ones in industry and construction have disappeared. The former type of jobs do not always have the better reputation, especially related to job content (work pressure, lack of autonomy, accidents at work).
• An almost open question is how are new jobs developing since the crisis? What types of jobs will be created when the economy picks up again?

Nevertheless, more ‘scarring’ is the prognosis. This conclusion is related to the duration of the crisis. Past academic research has shown that the related enduring job insecurities will have longer-term effects on health and well-being at work. Motivational problems and the risks of psychosocial problems will grow. It is also related to the fact that the younger generation entering the labour market today face great difficulties, particularly in those countries hard hit by the crisis. Studies of past recessions have shown that those who experienced unemployment early in their working life are more likely to be unemployed again in later years. Moreover, they are likely to earn less over their working life than those of their peers who find jobs more easily. The importance of the scarring effects depends on a series of factors including the loss of skills and work experience, and employers’ belief that spells of early unemployment signal low commitment to work or low productivity.

‘Not all in it together’

Throughout the report it emerges that the economic crisis has not had the same effect on all groups of workers.

The Eurofound report on wages and working conditions in the crisis showed that younger, low-skilled and low-paid employees already seem to be experiencing a greater decline in income and more job insecurity than other groups. The results of the current report confirm this pattern and, in particular, the ‘less choice’ impact on employment conditions.

Sector differences also play a role, especially in countries where industry and/or construction were hit harder by the crisis than other sectors. The impact on the crisis on the public sector (dealt with in a forthcoming EWCO report on Working conditions in central public administration in Europe (TN1303013S)) should also not be forgotten. In this current report, the focus is on general trends in the private sector.

The crisis situation is also characterised by an unfamiliar gender effect in the working conditions trends of many European countries. Sectors like manufacturing and construction have been harder hit by the economic crisis and are sectors with relatively more male workers. In many of the working conditions trends identified in this report, there has been a stronger change in the situation of male workers than of female workers. However, evidence is lacking in how this change in the burden of work between men and women in a household will also lead to a change in the responsibilities of combining work and family life.

Is there a crisis also for working conditions?

Until spring 2012, not all the countries covered by this report had been hit to the same degree by the economic crisis. In terms of GDP growth and the change in (un)employment levels, it is possible to distinguish groups of countries by the size of the economic crisis for the period 2008 to 2011. This ranking and grouping have been used analytically in the report.

Tables 12a and 12b summarise the most contrasting results in average working conditions trends between the countries hit harder (lower rows, country code in red) and confronted less with the crisis (upper rows, country code in green). Countries are ranked based on the extent of the economic impact of the crisis (average economic growth and change in unemployment rate). Reported changes in the relevant working conditions indicators between (mainly) 2007 and 2011 are presented. Colour is used to indicate a (strong) increase or decrease, with green considered a positive change and red a negative change. As such, this mapping visualises contrasting or stronger trends between countries where the economic impact of the crisis was smaller or larger (see also the correlation coefficient with the ranking of economic crisis). Only the indicators for which this contrast was clear are included in the table. Such a contrast can be read as additional proof that the crisis has a stronger effect on a particular working conditions indicator.

Notes: Countries least and most affected by crisis, % change country average, 2007–2011

Indicators:

Job insecurity = change in proportion between 2007 and 2012 of working people who think it is ‘very likely’ or ‘quite likely’ they will lose their job within the next six months (source: EQLS)

Change in the proportion of temporary/part-time employed (2007–2011) who give as reason for temporary/part time employment that they ‘could not find permanent/full employment’ (produce: LFS)

Change in net migration rate = percentage change between 2007 and 2011; net migration rate = immigration minus emigration (source: OECD)

Colour coding: dark green = relatively strong positive change; light green = relatively positive change; light red = relatively negative change; dark red = relatively strong negative change on average

Notes: Countries least and most affected by crisis, % change country average, 2007–2011

Indicators:

Work–life balance, work intensity, absenteeism, work accidents: dummy coding based on reported national figures; 1 = increase; 0 =stable; -1 =decrease (between 2007 and 2011)

Job satisfaction: percentage of employees ‘very satisfied’ (8–10 on 0–10 scale) (source: ESS, 2006, 2010)

Colour coding: dark green = relatively strong positive change; light green = relatively positive change; light red = relatively negative change; dark red = relatively strong negative change on average

Overall, Tables 12a and 12b show a strong picture (depicted in red) of more insecurity and less choice (involuntary part-time and temporary work). The country averages show the strong effect of the extent of the crisis on job insecurity and the involuntary choice of temporary or part-time work. There is also a greater chance that the net migration rate turns negative in countries where the size of the economic crisis was bigger. Employees of countries hit hard by the crisis seem to be confronted with a stronger impact on work intensity (proof comes from Ireland and Spain). The increase in job satisfaction is also less high in this group of countries and even negative (exceptions are Portugal and Spain). More positive averages are noted in relation to health-related factors such as absenteeism and accidents at work for these countries where the crisis was bigger.

Interesting outliers can also be detected in these country linkages of average trends in working conditions indicators and the extent of the impact of the economic crisis. Most interesting are possibly the sometimes specific results of Germany (fairing rather well in the crisis) and Spain (confronted more strongly by the crisis). Compared with other countries that did not do so badly in the crisis, in Germany a less strong decrease in involuntary temporary work can be seen. Job insecurity did decrease on average, and job satisfaction increased considerably. Spain saw a decrease in temporary and part-time work and an increase of low-wage work, which is in contrast with the average trends in countries such as Greece, Ireland and Latvia. In these latter countries, temporary and part-time work increased, and low-paid work decreased. The percentage of people very satisfied with their job increased also more considerably in Spain.

Impact of policy reforms

In a first reaction to the crisis, the introduction or expansion of short-work measures played a positive role in overcoming this first employment shock in a series of countries. In a second step, many countries have focused on new employment measures to get the unemployed back to work, wage restraint to restore business competitiveness and minimum wage policies. Training was another point of attention. Table 13 summarises those reforms not dealt with by the EWCO report on wages, the EWCO report on training and the EIRO report on the Impact of the crisis on industrial relations.

Notes: Wages and training are excluded from this table as they are dealt with by separate reports.

Norway has also introduced more strict rules for sickness leave benefits; not included: short work schemes, wage interventions and training policies

Source: National contributions

As different policy boxes in the report have shown, these further measures related to working conditions are very often related to a ‘flexibilisation’ of certain employment conditions such as:

• removing barriers on temporary contracts;
• easing of dismissal procedures (severance pay, termination period and so on);
• increasing working time flexibility (for example, rules on overtime or the reference period for calculating the performed working hours).

In addition, some countries also introduced more strict rules on obtaining sickness leave or sickness benefit.

Policies have been mainly introduced in those countries confronted with severe economic problems. These countries are joined by the Czech Republic and Poland. Ireland and Latvia, however, do not appear in this list even though they were also stronger hit by the crisis.

The following reforms seem to have had an impact on trends in working conditions since the crisis.

• The policies of short-time work schemes and/or working time accounts – expanded or installed in a range of EU countries in the first years of the crisis – had an additional impact on the reduction in the average working hours and reduced the unemployment shock.
• In the recent crisis period, policy interventions especially in the countries confronted with international fiscal bailouts have been targeting wage cuts. The prime example is Greece, where nominal average wages dropped 6.4% between 2009 and 2011.
• Nearly all EU Member States developed measures to increase training and skills upgrading of especially the more disadvantaged in terms of training participation. A series of mainly east European countries with traditionally a low level of temporary employment have introduced reforms to facilitate the use of temporary contracts. Since these reforms, temporary employment has been on the rise in the Czech Republic, Estonia, Lithuania and Romania.
• The changes in the legislation on employment protection in Greece, Italy, Portugal and Romania might explain part of the large growth in job insecurity in these countries. However, the national contributions of these countries do not describe any empirical studies that investigated this relationship.
• Pro-active strengthening of employability, a strong social safety net and clear rules on ‘job loss’ mitigate job insecurity. The job insecurity trend in the Nordic countries seems to prove this empirically.
• Hungary, Lithuania and Portugal liberalised their overtime rules. Hungary and Romania extended the reference period for calculating working time, facilitating ‘longer working hours’. These reforms do not seem to have caused deviations in working time trends.
• More stringent rules for sickness benefits introduced since the crisis in Estonia, Italy, Lithuania, Norway and Portugal have had the effect of reducing rates of sickness leave.

Policies for ‘sustainable work’?

Besides wage freezes, employment and training policies, the crisis reforms related to working conditions appear strongly geared towards flexibility for more working. Quality of work measures to promote better work are not seen as part of crisis policies.

Just before the crisis the EU embraced the flexicurity concept. It can be concluded that such an integrated approach, which combines flexibility reforms with security enhancements, has not come more into the forefront since the crisis – indeed the reverse is true. However, it remains interesting to look how east European countries like Bulgaria and Slovenia which embraced the flexicurity concept are developing these policies in the current period of economic hardship.

Looking to (part of) a Europe 2020 strategy that believes in an innovation union built not only on more but also better jobs, it is also important to note that the current crisis measures, when taken in relation to working conditions, do not focus on how to establish and expand ‘sustainable’ work. Nevertheless this report gives some policy pointers on this topic.

• Stress and frustration at the workplace are the mental image of crisis. In other words, continued and increased attention to psychosocial problems at work are key.
• Segmentation between a core and peripheral workforce has been enforced in recent years at many places. Flexibilisation means ‘more choice’ for the first and ‘less choice’ for the latter. Factors like skills and entrants to the labour market (youngsters and migrants) determine this segmentation.
• The different gender effect of the crisis – men are hit harder than women – leads to new challenges, but maybe also to new solutions to improve gender equality.
• Restructuring and downsizing are factors in stress and insecurity. However, they also create opportunities to transform job design and work organisation in order to strive for ‘active, smarter, sustainable work’ instead of ‘passive, strained, stressful work’ as the findings seem to suggest.

A crisis can be a period which stimulates people to reconsider traditional positions and strategies. It can also create an environment for radical innovations. The challenge to improve quality of work as one of the ways out of the crisis appears be remain.

Guy Van Gyes and Lise Szekér, HIVA-KU Leuven

Bibliography

Aronsson, G. and Gustafsson, K. (2005), ‘Sickness presenteeism: prevalence, attendance-pressure factors, and an outline of a model for research’, Journal of Occupational and Environmental Medicine, Vol. 47, No. 9, pp. 958–966.

Asfaw, A., Pana-Cryan, R. and Rosa, R. (2011), ‘The business cycle and the incidence of workplace injuries: evidence for the U.S.A’, Journal of Safety Research, Vol. 42, No. 1, pp. 1–8.

Bakker, A.B. and Demerouti, E. (2007), ‘The Job Demands–Resources model: state of the art’, Journal of Managerial Psychology, Vol. 22, No. 3, pp. 309–328.

Boone, J., van Ours, J.C., Wuellrich, J.-P. and Zweimüller, J. (2011), ‘Recessions are bad for workplace safety’, Journal of Health Economics, Vol. 30, No. 4, pp. 764–773.

Cichocki, S. and Tyrowicz, J. (2010), ‘Shadow employment in post-transition – is informal employment a matter of choice or no choice in Poland?’, The Journal of Socio-Economics, Vol. 39, (4), pp. 527–535.

Clark, A.E., Knabe, A. and Rätzel, S. (2010), ‘Boon or bane? Others’ unemployment, well-being and job insecurity’, Labour Economics, 17, (1), pp. 52–61.

Davies, R., Jones, P. and Nunez, I. (2009), ‘The impact of the business cycle on occupational injuries in the UK’, Social Science and Medicine, Vol. 69, No. 2, pp. 178–182.

De Cuyper, N. and De Witte, H. (2007), ‘Job insecurity in temporary versus permanent worker: associations with attitudes, well-being, and behavior’, Work and Stress: An International Journal of Work, Health and Organizations, Vol. 21, No. 1, pp. 65–84.

De Witte, H. (1999), ‘Job insecurity and psychological well-being: review of the literature and exploration of some unresolved issues’, European Journal of Work and Organisational Psychology, Vol. 8, No. 2, pp. 79–86.

De Witte, H. (2005), ‘Job insecurity: review of the international literature on definitions, prevalence, antecedents and consequences’, SA Journal of Industrial Psychology, Vol. 31, No. 4, pp. 1–6.

Debus, M.E., Probst, T.M., König, C.J. and Kleinmann, M. (2012), ‘Catch me if I fall! Enacted uncertainty avoidance and the social safety net as country-level moderators in the job insecurity-job attitudes link’, Journal of Applied Psychology, Vol. 97, No. 1, pp. 690–698.

Dew, K., Keefe, V. and Small, K. (2005), ‘Choosing’ to work when sick: workplace presenteeism’, Social Science and Medicine, Vol. 60, No. 10, pp. 2773–2282.

Dieckhoff, M. (2013), ‘Continuing training in times of economic crisis’, summary paper, Economic Crisis, Quality of Work and Social Integration: the European Experience in the Great Recession, European Social Survey and Organisation of Economic Co-operation and Development policy seminar, 21 January 2013, Paris.

Gallie, D. (2013), ‘Job control, work intensity and work stress’, summary paper, Economic Crisis, Quality of Work and Social Integration: the European Experience in the Great Recession, European Social Survey and Organisation of Economic Co-operation and Development policy seminar, 21 January 2013, Paris.

Gash, V. and Inanc, H. (2013), ‘Job insecurity and the peripheral workforce’, summary paper, Economic Crisis, Quality of Work and Social Integration: the European Experience in the Great Recession, European Social Survey and Organisation of Economic Co-operation and Development policy seminar, 21 January 2013, Paris.

Green, F., Mostafa, T., Parent-Thirion, A., Vermeylen, G., Van Houten, G., Biletta, I. and Lyly-Yrjanainen, M. (2011), ‘Is job quality becoming more unequal?’, conference presentation, Trends in Job Quality, 2011 Industrial and Labor Relations Review conference, 3 November, Ithaca, NY [to be published in a special issue of Industrial and Labor Relations Review].

Hansen, C.D. and Andersen, J.H. (2008), ‘Going ill to work – what personal circumstances, attitudes and work-related factors are associated with sickness presenteeism?’ Social Science and Medicine, Vol. 67, pp. 956–964.

Heponiemi, T., Elovainio, M., Pentti, J., Virtanen, M., Westerlund, H., Virtanen, P., Oksanen, T., Kivimäki, M. and Vahtera, J. (2010), ‘Association of contractual and subjective job insecurity with sickness presenteeism among public sector employees’, Journal of Occupational and Environmental Medicine, Vol. 52, No. 8, pp. 830–835.

Hijzen, A. and Venn, D. (2011), The role of short-time work schemes during the 2008–09 recession, OECD Social, Employment and Migration Working Papers No. 115, OECD Publishing, Paris, available at http://www.oecd-ilibrary.org/social-issues-migration-health/the-role-of-short-time-work-schemes-during-the-2008-09-recession_5kgkd0bbwvxp-en

Kahn, L.B. (2010), ‘The long-term labor market consequences of graduating from college in a bad economy’, Labour Economics, Vol. 17, No. 2, pp. 303–316.

Karasek, R.A. (1979), Job demands, job decision latitude and mental strain: implications for job design, Administrative Science Quarterly, Vol. 24, pp. 285–308.

Krillo, K. and Masso, J. (2011), Labour markets in the Baltic states during the crisis 2008–2009: The effect on different labour market groups, Working Papers of the Faculty of Economics and Business Administration, No 79, University of Tartu.

Luechinger, S., Meier, S. and Stutzer, A. (2010), ‘Why does unemployment hurt the employed? Evidence from the life satisfaction gap between the public and private sectors’, Journal of Human Resources, Vol. 45, No. 4, pp. 998–1045.

Naithani, P. (2010), Recession and work–life balance initiatives (152Kb PDF), The Romanian Economic Journal, Vol. 13, No. 37, pp. 55–68.

Prater, T. and Smith, K. (2011), ‘Underlying factors contributing to presenteeism and absenteeism’, Journal of Business and Economics Research, Vol. 9, No. 6, pp. 1–14.

Russell, H. and McGinnity, F. (2013), ‘Under pressure: the impact of recession on employees in Ireland’, British Journal of Industrial Relations, doi: 10.1111/bjir.12018.

Steiber, N. (2013), ‘Economic downturn and work motivation’, summary paper, Economic Crisis, Quality of Work and Social Integration: the European Experience in the Great Recession, European Social Survey and Organisation of Economic Co-operation and Development policy seminar, 21 January 2013, Paris.

Annex 1: Country groups and codes

Country groups

EU27 Current 27 EU Member States

NMS 12 New Member States that joined the EU in May 2004 (Cyprus, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovakia and Slovenia) and in January 2007 (Bulgaria and Romania)

Country codes

The order of the countries follows the EU protocol based on the alphabetical order of the geographical names of countries in the original language.

EF/13/19/EN