Employment Flexibility and Unequal Assessment of Fixed-Term and Permanent Employees' Training Needs in Europe

There is substantial disparity in training participation between permanent and fixed-term employees in Europe as studies consistently show a higher training participation rate among permanent employees (Cutuli and Guetto 2012; OECD 2002; Steijn, Need, and Gesthuizen 2006; Lucidi 2010; Forier and Sels 2003). Skills and knowledge acquired through company training are ‘stepping stones’ towards stable employment and good jobs (Gash 2008; Gebel 2010; Luijkx and Muffels 2008), it is important to understand the causes of this ‘training gap’.  Most studies focused on the individual-level determinants of training participation (Arulampalam and Booth 1998; Cutuli and Guetto 2012) and there is less research employers’ training strategies (but see O’Connell & Byrne 2012). This is an important gap to fill as participation in company training largely depends on employers’ provision of opportunities (Forier and Sels 2003). Moreover, no previous study has investigated the pattern of cross-country differences in organizational strategies of investing in the training of permanent and fixed-term employees. To better understand disparity in training participation this paper studies employers’ equal or unequal investment in assessing the training needs of fixed-term and permanent employees across different European countries.

As economic uncertainty and competition increases, fixed-term employment have become a viable (and global) human resource strategy of employers (Kalleberg 2000).  Combining mechanisms from human capital theory (Becker 1962; Acemoglu and Pischke 1998) and the theory of the flexible firm (Atkinson 1984, Lepak and Snell, 1999), I derive the hypothesis that a flexible hiring strategy of the firm is associated with more equal training investments in fixed-term and permanent employees. Employment protection legislation (EPL) regulate the dismissal of permanent employees to a varying degree across Europe (OECD 2013). I reason using the transaction cost theory that under ´inflexible´ labor policy conditions employers are more likely to unequally invest in the assessment of training needs in favor of permanent employees. Furthermore, I expect that under ´inflexible´ labor policy conditions the proportion of fixed-term employees has a smaller effect on the likelihood of unequal assessment of training needs.

I use data from 15 Western European countries in the European Company Survey of 2008 among companies with employees 50 or more (n=7811). As our interest is in firm’s strategic investment in fixed-term and permanent employees we restrict the dataset to organizations which during the year preceding the data collection and currently employ fixed-term employees (n=5446, 70 % of organizations). Assessment of training needs is measured by combining two questions of the survey: “Is the need for further training periodically checked in a systematic way in your establishment?” and the item “Is the need of employees with fixed-term contracts for further training periodically checked in a systematic way in your establishment?”. The dependent variable consist of 3 categories: no assessment of training needs (“no” on both questions, 13 % of organizations), equal assessment of fixed-term and regular employees (“yes” on both questions, 59 % of organizations), unequal evaluation of fixed-term and regular employees (“yes” on the first question, and “no” on the second question, 28 % of organizations). The manager-reported percentage of employees with fixed-term contract (mean=12.8 %, std =18) measures flexible hiring practices. Additional controls for firm size (8 categories), sector (1-digit NACE, 11 categories), percentage of workers with tertiary education, percentage female employees , employing temporary agency workers, and manager’s assessment of economic difficulties are included in the model. To assess the possibility of reverse causality - (lack of) training opportunities drive flexible labor practices -, we estimated an IV probit model of training assessment and instrument the presence of fixed-term employees with a variable whether employee representative bodies restriction of hiring fixed-term employees. We did not find evidence of reverse causality between training and hiring fixed-term employees. Building on the classification of Muffels and Luijkx (2008), we classified UK and Ireland in a ‘highly flexible’ category, characterized by comparatively lowest protection of permanent employees in Western Europe. The Netherlands, Denmark, Sweden and Finland are put in the ‘moderately flexible’ category. Germany, France, Belgium, Austria and Luxembourg are placed in cluster Continental inflexible, while Spain, Portugal, Greece and Italy are put in cluster Southern inflexible. Both Continental and Southern are characterized by higher labor protection, but Continental countries invest more in active labor policies and retraining, which leads to a higher-skilled pool of fixed-term employees and potentially less need of employer assessment of training needs. We estimated multinomial logistic regression with robust clustered standard errors to adjust for country clustering when analyzing the three training assessment decision scenarios (no assessment / unequal assessment / equal assessment). The low number of country units (n=15) does not allow multilevel estimations.

Preliminary results of this paper point out organizational and cross-country differences in the assessment of fixed-term and permanent employees training needs. Confirming our expectation, a higher share of fixed-term employees is related to greater equality in training. Employers in Continental and Nordic European countries are less likely to equally evaluate the training needs of fixed-term employees, compared to the more flexible UK and Ireland. Southern European countries do not differ from Continental and Nordic countries in this regard. We found that in Southern European countries the percentage of fixed-term employees has lower association with equal assessment than in other countries, which confirms our expectation that fixed-term employees have less training opportunity when policies higher employment protection for permanent employees. Although these regression results are cross-sectional, institutional and organizational-level mechanisms seem to interact in defining inequality in training opportunities. In more comprehensive analyses we aim to confirm the robustness of findings to other sources of cross-country variation that may drive training decisions of firms, for example differences in (vocational) educational systems (Vogtenhuber, forthcoming).