Background

In the early 20th century, Finland was still one of the poorest and most agrarian areas in Europe. In the post-war era, Finland caught up with the rest of the industrial world in a quantum leap from primary production to services and innovation-driven industries. In the latter half of the 20th century, as measured by gross domestic product (GDP) per capita, Finland became one of the most affluent societies in the world, surpassing, among others, the UK.

Policy choices were crucial. The development of welfare state mechanisms since the 1960s has provided the social cohesion necessary for the nation’s rapid modernization. In particular, systematic investment in education has made it possible to meet the demands of the changing industrial environment. In 2013, 46% of Finnish employees had tertiary-level education compared to a mere 10% in 1977 (Reference Sutela and LehtoSutela and Lehto, 2014: 17). Today, Finland ranks at the top in education and skills among the industrialized economies (Organisation for Economic Cooperation and Development (OECD), 2014).

Along with the development of welfare state mechanisms, trade unions have gained a strong position in the Finnish political decision-making arena, playing a key role in regulating the distribution of income and protecting workers’ rights. Although increasing emphasis is placed on local bargaining, major economic actions are usually taken after consultations among the government, trade unions, political parties and the business world. The rate of unionization is about 75%, one of the highest in Europe (Ministry of Employment and the Economy, 2012).

Finnish labour market regulation is best described as individually protective and collectively collaborative, albeit not overly restrictive from the employers’ point of view. There is no formal minimum wage legislation; instead, pay rates are set by legally binding industry-level agreements. The Working Hours Act, one of the key statutory mechanisms protecting employees’ rights, limits the maximum number of regular weekly working hours to 8 a day and 40 per week, but in practice industry-specific agreements are made through collective agreements. Weekly working hours do not typically exceed the standard 40hours (Reference Ojala and PyöriäOjala and Pyöriä, 2015).

In short, collaboration and collective negotiations are the guiding principles of Finnish industrial relations. Collective bargaining coverage in Finland is very high, including over 90% of all employees (www.worker-participation.eu/). The system is not without flaws, but it has provided a robust foundation for the development of job quality. In international comparisons, Finland, along with the other Nordic countries, seems to perform very well in terms of the quality of working life, as, for example, the comprehensive Eurofound (2012, 2015) surveys indicate.

In their analysis of the European Working Conditions Survey from 2005, collected by Eurofound, Reference Puig-Barrachina, Vanroelen and VivesPuig-Barrachina et al. (2014) compared the social distribution of precarious employment in the European Union (EU-27) countries. They based their analysis on a complex multidimensional construct including 11 indicators, including employment instability, low income level, insufficient protection, collective bargaining and training and low level of control over working time. The results speak in favour of the Nordic model of industrial relations: in Finland, Sweden and Denmark, work is less precarious than in other European countries.

Research objectives

As the brief outline above testifies, Finland has a relatively advantageous position considering working conditions and job quality. However, this is not to say the country is immune to the precarization of work. This is a global phenomenon transcending the borders of nation-states (International Labour Organization (ILO), 2012; Reference VoskoVosko 2010), although there is variation across capitalist regimes (Reference Oinas, Anttila and MustosmäkiOinas et al., 2012). Therefore, country-specific analyses are important. We ask the following research questions:

1. 1. What proportion of Finnish wage- and salary-earners are affected by the precarization of work?

2. 2. To what extent is a precarious labour market position associated with a low level of intrinsic job quality?

Our analysis is based on pooled data from Statistics Finland’s Quality of Work Life Surveys collected in 1984, 1990, 1997, 2003, 2008 and 2013. These are extensive studies with a very high response rate (68%–89%), involving between 3000 and 5000 people and covering the entire wage- and salary-earning population residing in Finland. The surveys have been conducted in the form of personal face-to-face interviews (Reference Sutela and LehtoSutela and Lehto, 2014).

The data set represents Finnish wage- and salary-earners aged 15–64 who regularly work more than 10hours a week. It does not provide us access to the hard core of the precariat, such as those whose income combines random sources and social security, but we do gain a picture of those who at the time of inquiry were in wage employment, even if this was on part-time or very short-term contracts.

The data set includes people employed under active labour market measures as well as those who have reluctantly accepted temporary or short-term employment. It allows us to make a comprehensive assessment of how the situation of wage- and salary-earners has developed over several decades, and furthermore to describe and identify the working conditions associated with a precarious labour market position.

Criteria of precarious employment

Defining and measuring precarious work is arduous. Precarious work is not a precise statistical category, as it includes a variety of aspects of employment relations and characteristics of jobs and working conditions (Reference KallebergKalleberg, 2014). Our purpose is to take into account the multidimensional nature of the phenomenon without making the empirical construct unnecessarily complex.

In the research literature, precarious work is typically associated with labour market uncertainty and risks, including atypical employment relationships, a high risk of job loss and poor prospects of employment if one is dismissed, and low earnings. These characteristics form the basis of our classification. This is not the only possible way to define precarity, but our choice is in line with earlier research (Reference KallebergKalleberg, 2011; Reference Rodgers and RodgersRodgers and Rodgers, 1989; Reference Tompa, Scott-Marshall and DolinschiTompa et al., 2007).

We operationalize the wage- and salary-earning precariat using the following five criteria, which reflect both objective and subjective factors related to a weak labour market position:

1. 1. Atypical employment relationship (temporary labour and agency workers);

2. 2. Realized labour market risk (at least one spell of unemployment in the past 5years);

3. 3. Fear of labour market risk (is layoff, dismissal and/or unemployment considered a threat?);

4. 4. Poor prospects of employment (assessment of chances of finding a new job in the open labour market);

5. 5. Low earnings (lowest pay quartile).

Our first criterion is an atypical employment relationship, defined as comprising those working on temporary contracts since the 1980s as well as agency workers since 2008 (the year when this item was included for the first time in the Statistics Finland survey). We exclude part-time employment from this classification because it cannot be simply equated with precariousness (Reference KallebergKalleberg, 2014). For most people in Finland, part-time work is a voluntary choice based on open-ended employment contracts (Reference Sutela and LehtoSutela and Lehto, 2014).

Instead, we focus on atypical employment that is linked to job insecurity. According to Reference KallebergKalleberg (2014), the priority in measuring precarious work should be the extent to which work is insecure, uncertain and unstable. Earlier empirical research has identified uncertainty as one of the most important perceived work-related threats (Reference Cheng and ChanCheng and Chan, 2008; Reference Sverke and HellgrenSverke and Hellgren, 2002).

Our second criterion is realized labour market risk, that is, whether the respondent has been out of work or laid off once or more during the 5years preceding the survey.

Our third criterion is the fear of labour market risk. This is the sum of three risk factors (whether being laid off, dismissed and/or made redundant is perceived as a threat) that we form into a dichotomous variable (no threats vs at least one perceived threat). Although this fear is subjective, evidence suggests that employees are quite accurate in assessing the likelihood of labour market risks materializing in their current job (Reference GreenGreen, 2006).

Our fourth criterion is poor prospects of employment. This is measured by the question, ‘What do you think would be the likelihood of you finding a new job: good, reasonable, or poor?’ The response is deemed to reflect precarity if the respondent feels he or she has poor chances of finding a new job in the open labour market.

Finally, we include low earnings in our classification, as this is a major factor of uncertainty for employees (Reference OlsthoornOlsthoorn, 2014). By low earnings, measured by the question, ‘What is your monthly gross pay in your main job before tax?’ and standardized separately in each annual data set, we refer to the lowest income quartile. This choice is in line with the European Commission’s (2004: 60) analysis of precarious employment.

Respondents meeting three or more of these five criteria are considered to be in a precarious position. Most wage- or salary-earners have been affected by uncertainty in one form or another; very few indeed meet none of these criteria (see Table 2). So we define precarity essentially in terms of the accumulation of uncertainty. These measures are valid and reliable indicators of insecurity, in that their intercorrelations are fairly low, which means they are independent of one another. If they had correlated closely, we would be measuring one and the same dimension and fail to uncover the depth and complexity of the respondent’s experience.

Intrinsic job quality

In addition to departing from the norm of standard and secure employment relationships, precarious work is often related to poor job quality (Reference KallebergKalleberg, 2012, Reference Kalleberg2014). Job quality is a broad concept encompassing various individual, social and organizational aspects. Reference Green and MostafaGreen and Mostafa (2012) identify the following key dimensions: earnings, job security, career prospects (employability) and intrinsic job quality. As our classification of precarious work includes the first three dimensions, we focus on intrinsic job quality, a concept used by Green and Mostafa to refer to (1) skill requirements and discretion over daily job tasks, (2) the social and (3) physical work environment and (4) work intensity.

Reference Green and MostafaGreen and Mostafa (2012) define skill requirements in terms of job complexity and the individual employee’s discretion, which help strengthen the employee’s position in the labour market. Autonomy supports control over the job and well-being in the workplace, as does variety in work, and these factors strengthen the individual employee’s commitment to the job (see also Reference Gallie and GallieGallie, 2008). We measure five items: socio-economic status as a proxy measure for skill level, training provided by the employer, opportunities for self-development, variety in work and autonomy.

In the social environment of the workplace, support from superiors and colleagues is an important protective factor for employee well-being (Reference Green and MostafaGreen and Mostafa, 2012). In addition to such support, we are interested in precarious employees’ assessment of the climate and sense of solidarity in the workplace, as well as in their mental workload.

Physical job safety and the absence of health hazards in the workplace have always been crucial job quality factors, especially in earlier decades when mechanisms of labour market regulation were being developed (Reference Green and MostafaGreen and Mostafa, 2012). Consequently, advanced industrial countries have seen a declining incidence of occupational accidents and physical health hazards (Reference GreenGreen, 2006); Finland has very high standards of occupational safety. In our analysis, we take into account 15 physical workload factors (a sum variable).

Another major focus in studies on job quality has been work intensity. Research indicates that the intensity of work has been increasing in advanced industrial countries (Reference De Bustillo, Fernández-Macías and AntónDe Bustillo et al., 2011; Reference Green and MostafaGreen and Mostafa, 2012). In Finland, however, it seems that the ongoing financial crisis contributed to reducing work intensity between 2008 and 2013 (Reference Sutela and LehtoSutela and Lehto, 2014: 69–74). We have two measures to describe experiences of work intensity. Table 1 gives a detailed description of the indicators reflecting intrinsic job quality.

Table 1. Indicators describing intrinsic job quality.

In the analyses, we control for the factors that earlier research has shown to be associated with uncertainty of work (e.g. Reference GreenGreen, 2006; Reference Green and MostafaGreen and Mostafa, 2012). In addition to gender, age, and education/socio-economic status, we control for the background variables of variety in work (monotonous vs varied job content) and trade union membership (the high degree of unionization in Finland affords comprehensive protection to wage- and salary-earners).

Field of employment and place of residence are also controlled for: this is important because the Finnish labour market is characterized by high structural and regional differentiation (Ministry of Employment and the Economy, 2010). We divide the country roughly into two parts: the densely populated southwest (the triangle formed by the three biggest cities) and rural Finland.

Furthermore, we standardize whether or not the respondent is studying while working in gainful employment. Students may be more inclined to think they have good chances of employment and they are not yet burdened by prolonged experiences of uncertainty. We also control for regular working hours as a continuous covariate since the most recent data available suggest that people in precarious positions work a couple of hours less per week than wage- and salary-earners on average. Finally, we control for the time point (year) because the data set is pooled. Unfortunately, we could not control for ethnicity because the data contain no such information.

Precarious employment and economic fluctuations from the 1980s to the 2010s

As Table 2 shows, the proportion of precarious wage earners has varied according to economic cycles. The figure was at its lowest at 8% in 1990, just before Finland plunged into a deep recession, and at its highest in 1997, when the proportion of precarious workers climbed to 18%. Finland took longer than the rest of Europe to shrug off the recession, which explains the high figure for 1997. The financial crisis that reached Finland in late 2008 had less impact than the 1990s great recession, although a rebound is still not in sight. While the proportion of precarious wage- and salary-earners has increased, overall the situation in 2013 hardly differs from that 10years earlier, when the economic outlook was strong.

Our assessment of the prevalence of precarity suggests that there have been no radical changes over the past three decades. The one notable exception is 1997. The 1997 data set represents a more selected sample of wage- and salary-earners than usual because of the high level of unemployment. In the latter half of the 1990s, the structure of labour demand changed much more rapidly than the structure of the labour force itself. As a result, even a high level of education and a permanent contract did not provide protection against redundancy (Reference Kalela, Kiander and KivikuruKalela et al., 2001).

Our assessment of the prevalence of precarity suggests that there have been no radical changes over the past three decades. The one notable exception is 1997. The 1997 data set represents a more selected sample of wage- and salary-earners than usual because of the high level of unemployment. In the latter half of the 1990s, the structure of labour demand changed much more rapidly than the structure of the labour force itself. As a result, even a high level of education and a permanent contract did not provide protection against redundancy (Reference Kalela, Kiander and KivikuruKalela et al., 2001).

The criteria listed in Table 2 show no major changes from 1984 to 2013, with the exception of fears of a personal labour market risk materializing (+18percentage points). As the financial crisis has persisted, increasing numbers have expressed fears for their job. Fears and risks concerning one’s own labour position are cyclically sensitive. Uncertainty tends to increase in bad times, and confidence during good times. In Table 2, the macroeconomic situation seems most clearly reflected in perceived prospects of re-employment. These assessments were the most optimistic ahead of both recessions, in 1990 and 2008.

In 1984, assessments of employment prospects were surprisingly low, close to 1997 figures, possibly because of the lower level of labour market mobility in the 1980s. This probably also explains why despite the current financial crisis, people have not lost faith in their chances of finding a new job, although the proportion who regard their employment prospects as poor increased somewhat (+5percentage points) in the wake of the recession from 2008 to 2013.

Overall then, there was increasing precarity in the Finnish labour market in the 1990s, but in the 2000s the situation returned to normal. During the financial crisis of 2008–2013, the proportion of precarious wage- and salary-earners increased, but uncertainty did not reach the same level as during the previous recession.

The accumulation of precarity?

We move on now to discuss the accumulation of uncertainty in certain groups. The odds ratios in Table 3 allow us to infer which background factors explain accumulating uncertainty. Cyclical fluctuation, which can be demonstrated by controlling for the time point, is associated with the precarization of work. However, the accumulation of uncertainty is most clearly reflected by individual factors, such as gender, being young, and having a low level of education, and structural factors, such as place of residence and field of employment.

The impact of place of residence is as expected: living outside urban Finland increases the risk of a precarious labour market position. As in other European countries, urban centres attract new enterprises and a talented workforce, powering innovation and economic growth, whereas rural areas are regressing.

The impact of field of employment has more intricate implications linked to atypical employment relationships and the risks associated with unemployment. On the one hand, employment in public services (education and research, and health care and social work, in particular) indicates uncertainty, largely attributable to the high prevalence of fixed-term contracts in these fields of employment. This is a distinctive feature of the Finnish labour market: temporary contracts are most common among highly educated public sector employees. In this sector, however, temporary contracts are often renewed and perceived prospects of re-employment are good.

On the other hand, uncertainty is no stranger to workers employed in primary production (agriculture etc.), construction and manufacturing. In these sectors, realized labour market risk and fear of unemployment are the most significant factors predicting precariousness. In recent years, cyclically sensitive fields of employment, such as construction and manufacturing, have suffered from the ongoing financial crisis, while public sector jobs have been more secure.

Gender, age and education proved to be the most important individual factors predicting precariousness. Women have a higher risk of being in a precarious labour market position than men, based on low earnings and poor prospects of employment. Being young and having a low level of education also increase the risk of a precarious labour market position. Studying has the opposite effect: acquiring an education is an intermediate phase that includes finding one’s own place in the labour market and drifting through short-term jobs.

We should be cautious not to draw far-reaching conclusions from the association observed between youth and precarity. The association between age and a precarious labour market position is no longer statistically significant among wage- and salary-earners over 35years old. Youth is a time of transition and exploration, and this has always involved an element of uncertainty.

Precarious work and its relation to intrinsic job quality

What kinds of working conditions are associated with a precarious labour market position? In Table 4, we assess how precarious wage earners compare with other employees in terms of the features of intrinsic job quality described in Table 1. In terms of skill use and discretion, the social and physical work environment and work intensity, Table 4 compares the means for precarious and other wage- and salary-earners.

Our key observation is that both privilege and disadvantage accumulate in the labour market. More surprisingly, among the indicators we have chosen to use here, there is not a single characteristic of precarious work that substantially offsets a weak labour market position. Precarious workers are in a worse position than others according to all other indicators except for mentally demanding tasks and work intensity. The effect of adverse work environment factors (heat, cold, etc.) and support from colleagues proved to be statistically non-significant.

The negative association observed between precarity and skill use and discretion is particularly pronounced and problematic. Precarious workers have less on-the-job training, variety in work and autonomy, as well as fewer opportunities for self-improvement than other employees. Yet these factors provide important protection against stressors at work, and the development of independent skills in particular is key to determining the employee’s future labour market position (Reference GreenGreen, 2006).

Furthermore, a precarious labour market position indicates a social climate that is less supportive of well-being in the workplace, for instance, in terms of support from one’s superiors or collegial respect and appreciation (i.e. being a valued member of the work community). Earlier studies have found that the workplace climate and support from superiors provide protection against the stresses and experiences of injustice resulting from factors of insecurity such as fixed-term employment (Reference LiukkonenLiukkonen, 2012).

In other words, the accumulation of work-related uncertainty is associated with weaker than average intrinsic job quality. In this respect, the criticisms raised in the precarity debate are well placed.