Work safety in the context of temporary employment: The spanish experience.(Statistical Data Included)

CATALINA AMUEDO-DORANTES (*)

A rise in work injuries in Spain that has coincided with increased employment flexibility has fired debate among unions, employers, and policy-makers. This paper examines the relationships among temporary work, working conditions, and the likelihood of work accidents and illnesses. An analysis of employee-level and establishment-level data from a 1997 survey conducted by the Spanish Department of Labor suggests that, consistent with a claim made by unions, the single most important determinant of the likelihood of work-related injury and illness is working conditions, not education or tenure. As a result, although temporary workers exhibit higher work injury and illness rates than permanent workers, they exhibit a lower likelihood of work injury and illness than permanent workers once the analysis controls for a given set of working conditions.

Work injuries in Spain have been on the rise since 1994. Part of this increase may be attributable to the growth in the employed population. (1) However, the number of accidents resulting in lost work-time as a fraction of the working population--the incidence rate--has also been rising. Between 1996 and 1997, the incidence of work-related accidents increased by 9.4% (Ministerio de Trabajo y Asuntos Sociales 1998). The growth of work-related illnesses was also rapid (Ministerio de Trabajo y Asuntos Sociales web page). Overall, the number of work-related injuries and illnesses grew during the second half of the decade despite the approval in 1995 of Law 31 for the Prevention of Work Risks, which contained strict safety rules for the workplace.

The sustained growth of job injuries is worrisome given its impact on workers' wellbeing and its high cost to firms and the economy as a whole. (2) Spanish unions have charged that the rise in work accidents is linked to conditions endured by the large fraction of workers on fixed-term contracts (Garcia 1999; Serrano Pernas 1999). Indeed, the incidence rate of work accidents among workers with fixed-term contracts has at times been more than double that of workers with open-ended contracts during the past decade (Ministerio de Trabajo y Asuntos Sociales web page).

Why is this the case? Do permanent and temporary (3) workers enjoy different degrees of workplace safety, or is the increase in the work injury rate linked to temporary employees' personal characteristics and job history? Can we still draw a link between the likelihood of work injuries and type of work contract once we control for working conditions, workers' human capital, and general establishment characteristics? What factors are driving the likelihood of work accidents and illnesses? The tripartite discussions among unions, employers' organizations, and the government on work safety have proceeded as if the effects of employment flexibility and the safety measures adopted by the Law 3l/l995 were well documented. However, our knowledge about work injuries is limited to descriptive analysis using data from either the Labor Force Survey (Encuesta de Poblacion Activa, or EPA) or the Work Accidents Survey (Encuesta de Accidentes de Trabajo, or EAT), which provide little or no information on the employee's work ing conditions and safety measures adopted by the establishment.

This study is an attempt to fill this void in the literature and address the aforementioned questions using establishment- and employee-level data from the 1997 Encuesta Nacional sobre Condiciones de Trabajo (ECNT). The ENCT is a one-time survey carried out in 1997. (4) It collected workplace safety information in two questionnaires administered separately to employers and employees in all economic activities except agriculture and mining.

While I rely on Spanish data, the richness of the ENCT allows this study to contribute to the international debate over work safety and new employment relations in three ways. First, the survey's coverage of multiple economic activities makes it possible to extend the existing, primarily industry-specific analyses of the relationship between temporary employment and work injuries to a variety of economic sectors. (5) Second, combining employee and employer data makes it possible to better address the link between contract type and work safety. The employer questionnaire asks the employer about the nature of the establishment and the adoption of various safety and preventive measures. The employee survey also collects information regarding the implementation of a variety of government-regulated safety measures. Most notably, however, it solicits information about the employee's personal and job characteristics, specific working conditions that affect the occurrence of job injuries--such as the manipulation of dangerous machinery or toxic products--and the employee's involvement in work safety decisions. In analyzing the determinants of the employee's likelihood of work injury, I can use the data on personal and job characteristics to separate the effects of working conditions from the effects of education, tenure, work experience, occupation, and type of contract. The collected information on the employee's specific working conditions is also of interest, since those conditions might change by occupation within the establishment regardless of the general safety measures adopted by the firm. Third, having information on employee involvement in the firm's work safety decision-making allows an examination of the unexplored relationships between temporary employment, employee involvement, and work safety (Barker and Christensen 1998b).