In recent years employers and policy makers have considered labour market flexibility as a means of improving workers’ performance and adaptability to technical change and increasing globalisation.¹ Alongside these changes within production, employment conditions have become less stable. However, very little research has been done to analyse the impact of flexible employment on people’s health and life conditions. This knowledge is likely to provide some insights into equity considerations that would complement the arguments of those who promote greater flexibility in the labour market based on productivity reasons.

Previous research on job insecurity and health has been largely based on the analysis of perceived insecurity and positive relations with poor psychological and physical health have been found.^2–7 This approach, however, has some limitations because feelings about job insecurity are to some extent affected by aspects other than the objective contractual arrangements.⁸ For example, Orpen⁹ reported that although perceived flexible employment was related to anxiety and depression among all employees, these mental health indicators did not differ between employees in jobs objectively secure or insecure. Actually, when both dependent and independent variables are subjective and self reported, personality may influence both of them and associations can be overestimated because of the sharing of a common variance.

Recently a handful of studies have explicitly analysed the association between different forms of unstable contractual employment arrangements and mortality¹⁰ or other health outcomes,^11–15 but their findings are inconsistent. Although the existence of differences between geographical contexts, with different cultural backgrounds and labour policies cannot be ruled out,^13,16 there are a number of methodological limitations that could also explain these contradictions. An issue that the research in this area has to face is that of reverse causation, whereby it is not the case that the experience of job instability leads to poor health status but the opposite, that is, people with poorer health are more likely to work in instable job arrangements. For example, it has been shown that among temporary workers good self perceived health status, not having psychological distress, and a non-sedentary life style are associated with achieving a permanent job contract.¹⁷

On the other hand, many studies have been based on one health indicator but the effect of flexible employment can differ depending on the wellbeing indicator used. For example, it can affect workers’ social health by limiting partnership formation or the decision of being a parent. It has been shown that holding a job is an important predictor for union formation and parenthood.¹⁸ However, the lack of contractual security can be an additional limitation for the assuming of family responsibilities. Employees expect their contract to finish within a given time period and it is difficult to foresee their future labour market career and therefore also to plan ahead.

At this stage of research it is also important to find out if certain social groups are more or less vulnerable to the damaging effects of flexible employment. For instance, a criticism about the sex imbalance in the existing literature about health and contingent work has been raised.¹⁹ Whereas some studies have not found sex differences in the impact of flexible employment on health,²⁰ others have reported less noticeable differences for women^21,22 and some have simply adjusted the analysis for sex.^11,13 Moreover, even though it has been reported that social class modifies the effect of unemployment on mental health,²³ as far as we know, this issue has not been explored in the research about flexible employment. Less qualified workers could be more vulnerable to flexible employment because of their lower employability and less power to negotiate their employment conditions.

The objectives of this study are (1) to analyse the impact of flexible employment on mental health and job dissatisfaction; and (2) to examine the constraints imposed by flexible contractual arrangements on men’s and women’s partnership formation and people’s decision to become parents. For the two objectives the potentially different patterns by sex and social class will be explored.

METHODS

Data

The data for this study were derived from the 2002 Catalonian health survey (CHS), a cross sectional study based on a representative sample of 8400 members of the non-institutionalised population of Catalonia, a region in north eastern of Spain with about seven million inhabitants. The survey gathered self reported information on morbidity, health status, health related behaviours, as well as sociodemographic data. Participants were randomly selected via a multiple stage random sampling strategy. In the initial sampling stage, municipalities were selected from each of the eight regions of Catalonia (or municipal districts in the case of the Barcelona City health region) according to their population size. In each of the resulting strata cluster random sampling involving proportional probabilities and municipality (or district) weights, was used to select people for participation. Only 12.7% of respondents were replaced as a result of refusal or absence. Full details of the survey have been reported elsewhere.²⁴

For the purposes of this study we selected a subsample of all salaried workers aged 16–64. To avoid possible reverse causation effects, we excluded from our analysis those people reporting a longstanding limiting illness (LLI) in the 12 months preceding the survey, 9.3% of men and 11.5% of women. LLI was asked with the question “During the past 12 months have you had any trouble or difficulty for gainful employment, housework, schooling, studying, because of a chronic health problem (that has lasted or it is expected to last three or more months)?” The final sample under analysis was composed of 1474 men and 998 women.

Dependent variables

• Poor mental health was measured with the 12 item version of the general health questionnaire (12-GHQ),²⁵ included in the CHS questionnaire. This is a screening instrument widely used to detect current, diagnosable psychiatric disorders. It focuses on breaks in normal functioning rather than on lifelong traits; therefore it covers personality disorders or patterns of adjustment when these are associated with distress. Here we used a two point scoring method, rating a problem as absent (0) or present (1) according to the method recommended by the GHQ. Responses were summed and those scoring 3 or more were classified as cases.²⁶

• Job dissatisfaction was measured with the question “All in all, how satisfied would you say you are with your job?”, with four categories, from “very satisfied” to “very dissatisfied”. This question was dichotomised by combining the categories “dissatisfied” and “very dissatisfied” to indicate job dissatisfaction.

• Limitations in partnership formation and parenthood:

  • Marital status was categorised as never married compared with married or cohabiting. For this analysis, previously married people were excluded.

  • Parental status was categorised as living with children at home or not. This analysis was restricted to people married or cohabiting.

Flexible employment

Flexible employment was assessed through the type of contract. The question had four categories: permanent, fixed term contract, non-fixed term temporary contract, and working with no contract. The three latter types of contracts were considered as of objective flexible employment.

Measurement of occupational social class

Occupational social class, assigned according to the respondent’s current occupation was measured with a widely used Spanish adaptation of the British classification.²⁷ Class I includes managerial and senior technical staff and freelance professionals; class II, intermediate occupations and managers in commerce; class III, skilled non-manual workers; class IV, skilled (IVa) and partly-skilled (IVb) manual workers; and class V, unskilled manual workers. Because of the low number of people in some categories, the six original social classes were collapsed into two broad classes: non- manual (I and II, III) and manual (IVA-IVB-V).

Data analysis

Multiple logistic regression models were fitted to calculate adjusted odds ratios (aOR) and 95% confidence intervals (CI). Models were separated for sex and social class and adjusted for age. Goodness of fit was obtained using the Hosmer Lemeshow test.²⁸ The analysis included the weights derived from the complex sample design.²⁹

RESULTS

General description of the population by sex and social class

Table 1 shows the general description of the population. Whereas mental health status was better among men, no class differences were found either for men or for women. Job dissatisfaction was higher among manual female workers. Working with temporary contracts was more frequent among manual workers and among women. Moreover, more than 8% of female manual workers were employed with no contract. Being married or cohabiting was more frequent among non-manual male workers, whereas female non-manual workers were more likely to remain single. This pattern was consistent with that of living with children at home.

Poor mental health, job dissatisfaction, and type of contract by sex and occupational social class

Table 2 shows the association of type of contract with poor mental health and job dissatisfaction. Working with fixed term contracts was not associated with poor mental health status in any sex and social class combination. Among non-manual female workers and manual male workers non-fixed term temporary contracts were positively associated with poor mental health (aOR = 3.87; 95% CI = 1.52 to 9.85 and aOR = 4.30; 95% CI = 1.96 to 9.44 respectively). Among manual workers, both men and women, people with no contract were more likely to report poor mental health status (aOR = 6.34; 95% CI = 1.89 to 21.22 and aOR = 3.27; 95% CI = 1.54 to 6.94 respectively).

The pattern of association between job dissatisfaction and type of contract was different than that for mental health. As with mental health, among manual workers working with no contract was positively associated with job dissatisfaction both among men (aOR = 3.88; 95% CI = 1.19 to 12.68) and women (aOR = 4.91; 95% CI = 2.25 to 10.71). Additionally, among non-manual men those with fixed term contracts were more likely to report job dissatisfaction (aOR = 2.72; 95% CI = 1.16 to 6.37).

Limitations in partnership formation and parenthood

Table 3 shows the association of type of contract with marital and parental status. In all groups except manual female workers all forms of non-permanent jobs were associated with non-partnership formation although few associations were statistically significant. Limitations in parenthood were only found among men, both manual and non-manual, with fixed term temporary contracts (aOR = 1.97; 95%CI = 1.10 to 3.53 and aOR = 2.94; 95%CI = 1.16 to 7.47 respectively).

DISCUSSION

Our study produced three main findings: (1) whereas non-fixed term contracts and working with no contract were associated with poor mental health status, no association with fixed term temporary contracts was seen; (2) the effect of flexible contractual arrangements, other than fixed term temporary contracts, on mental health was higher among less privileged groups (women and manual male workers); and (3) the impact of flexible employment, either fixed term or non-fixed term contracts, in family formation was more pronounced among men.

Sex differences in health and employment conditions

As expected, mental health status was poorer and flexible employment more common among women and manual workers.^30–32 The more frequent type of non-permanent contract was fixed term temporary contracts. In both sexes, flexible employment increased among manual workers. It is important to note that 8.4% of female manual workers had no contract. Most of them (63%) worked as cleaners or carers of dependent people in domestic settings. These findings emphasise the need of increasing research into traditional female jobs as well as the impact of flexible employment in such jobs

Impact of flexible employment on mental health and job dissatisfaction

Several studies have shown the negative impact of job insecurity on mental health.^7,11,33 However, we have found no relation between working with fixed term temporary contracts and mental health or job dissatisfaction. In contrast with our result, a higher risk of poor self perceived health status among German workers with fixed term temporary contracts has been reported.¹³ There are several reasons that could explain these contradictory results. The health indicator analysed in the study of Rodriguez,¹³ self perceived health status, is different, reverse causation was not controlled for and, finally, differences between geographical contexts can exist.¹⁶ Actually, in that study no association between fixed term temporary contracts and poor self perceived health status was found in Britain. That is consistent with our results and with another study based on the British household panel survey.¹⁴

To the extent that work related insecurity limits workers’ perception of control over their lives, when this is a permanent situation it is likely to be related to chronic stress disorders.³⁴ Unfortunately, the category of fixed term temporary contract is too heterogeneous, including workers perceiving their situation as a transition to a stable situation, as for example those in the initial steps of their working life, and others that after many temporary contracts for a long period, often in several companies, are unable to get a permanent contract and live with a high uncertainty and concern about their working future. This heterogeneity could be a reason for not finding any association between working with fixed term temporary contracts and mental health. In future research, new variables should be considered to capture situations of precariousness, often associated with flexible employment arrangements, such as the uninterrupted time working with non-voluntary temporary contracts, the number of companies or workplaces where the person has been working in a defined period of time or the number of contracts in the past months, for example.

We have found some evidence of poor health outcomes in those forms of flexible employment more likely to be associated with poor employment conditions, such as working with non-fixed term temporary contracts and being employed with no contract. These results are consistent with those of Bardasi et al¹⁴ who did not find any association between fixed term temporary contracts and poor mental health but, in both sexes, found some evidence of poor mental health among workers with casual contracts. Moreover, we have also examined the role of social inequalities and we have additionally shown that the negative impact of these forms of contractual arrangements have a higher impact among less privileged workers—that is, women and manual male workers. These results challenge the view that the health effects of flexible employment emerge in a universalistic manner through individual perceptions that bear little relation to social structure.³⁵

In Spain the proportion of employees working with temporary contracts is almost three times that of the EU-15.³⁶ There is a general concern about the fraud in the use of these kinds of contractual arrangements, and primarily about the abuse of non-fixed term contracts as an easy form for employers to adjust the size or composition of their work forces to changing economic conditions. Job flexibility is used as a means of reducing labour costs and increasing control over workers. Although these contracts were originally created for use exclusively in activities of uncertain and limited duration, they are currently associated with precarious work. For example, large construction firms undertake different projects each one easily identifiable and it is the sum of all of them that guarantees the continuity of their economic activity. However, for each project the employer can make non-fixed term contracts and even legally make successive contracts to the same persons for different projects.

In our study 63% of manual male workers with non-fixed term contracts were employed in the construction sector. A significant proportion of persons worked more than 40 hours a week (46% compared with 26% in persons with permanent contracts). It is probable that people with contracts of unknown duration accept hard working conditions and different forms of exploitation, such as long hours, to keep their job (actually, it has been reported that employees with temporary contracts are less likely to have sickness absences).^11,37 This can be even more frequent in the Spanish labour market with a high unemployment and growing rates of immigrant workers rate. These results suggest that it is not only the uncertainty about job future that relates to poor mental health status but also the poor working conditions associated with some temporary contracts. However, this is just a speculation that deserves further attention in future research.

The pattern of association of type of contract with job dissatisfaction was different than that for mental health status. However, consistently with the results for mental health, those manual workers with no contract, either men or women, were more likely to report job dissatisfaction. On the other hand, among men, non-manual workers with fixed term contracts showed a higher risk of job dissatisfaction. No association was found with non-fixed term temporary contracts.

Temporary work and limitations in partnership formation and parenthood

As hypothesised, workers with temporary contracts, either fixed term or non-fixed term, were less likely to decide partnership formation and entry into parenthood, but this association was more pronounced among men. It is important to notice that whereas mental health was not related to fixed term temporary contracts, the delay in partnership formation and parenthood is primarily related to this type of contract. This finding suggests that whereas working in a temporary job is not a cause of poor mental health in itself and that more attention should be paid to poor working conditions associated with flexible employment arrangements, uncertainty about the future related to temporary work delays family formation. Married men with fixed term temporary contracts, either of high or of low social class, were less likely to have children. Moreover, among non-manual male workers these contracts were also associated with a lower probability of being married. Conversely, among manual male workers not being married was related to working with no contract.

In Spain holding a job is an important predictor for union formation and parenthood among men.³⁸ In such a country with a strong version of the male breadwinner model, long term and full time employment for men are considered necessary to consolidate an economic basis that is regarded as a necessary prerequisite for these transitions. Among women an association between not being married and working with a non-fixed term temporary contract was only found among non-manual workers. This finding could be explained, among other reasons, by the higher attachment of women of higher education levels to the labour market.³⁹ Besides the impact of flexible employment in shaping the capacity of both women and men to realise their potential for health, these results illustrate its contribution in the low birth rate in Spain, the lowest in the EU-15 after Ireland,⁴⁰ and emphasises the need for political interventions to reduce the abuses of temporary forms of employment arrangements.

Limitations

This study is limited by its cross sectional design. Although we cannot entirely rule out the possibility that poor mental health causes flexible employment, we reduced the reverse causation effect by excluding those workers who reported a LLI the past 12 months. Thus, it seems reasonable to think that, in most cases, flexible employment was not caused by a health problem. Therefore, we have studied a selected population of “healthy” workers and still we have found a poorer mental health status, higher job dissatisfaction, and limitations in living arrangements in selected groups of workers with flexible contractual arrangements. Moreover, we have used an objective measure of flexible employment, the type of contract, therefore avoiding an overestimation of the associations between subjective dependent and independent variables due to sharing part of the same variance.

We are also limited by the use of secondary data. So, we have considered living with children at home as a proxy of parental status, but in some cases children could have left home. However, the probability of leaving home does not seem to be related to the type of contract of parents or to be related in a different way depending on whether it is the mother or the father who has a temporary contract. Moreover, one of the main characteristics of Spanish society is the delay of the young in leaving the parent’s home because, among other reasons, of difficulties of the labour market and to the high cost of dwellings.⁴¹

Finally, another limitation is the lack of a sufficient number of people of some categories that has prevented us from analysing them. Moreover, partially because of the same reason, we could not examine the role of part time work, frequently considered as a form of atypical or precarious employment,¹⁹ because it is still rather uncommon in Spain (only 3.6% of men and 14.9% of women in the sample had these kind of contracts) and it is highly correlated with non-permanent contracts (whereas among full time workers the proportion of people with flexible contractual arrangements was 22% and 28% in men and women respectively, among part time workers it increased to 56%, with no sex differences).

Conclusion

We have shown that the impact of temporary work on mental health is related to those forms of employment arrangements that are potentially associated with precariousness, working with non-fixed term contracts, and with no contract, but not with fixed term contracts. We have also shown that this impact is restricted to less privileged workers such as women and manual male workers. Finally, we have found that, primarily among men, temporary work, either fixed term or non-fixed term, is associated with limitations in partnership formation and the decision to become parents.

Future research should characterise the forms of precariousness associated with flexible contractual arrangements, such as the number of contracts or the firms or workplaces where workers have stayed in the past months and the uninterrupted time working with temporary contracts, as well as the potential sex and social class differences.